Quantum Corporation (QMCO) Earnings Call Transcript & Summary

Thank you, Randy. I'd like to welcome everyone to Quantum Analyst and Investor Day. This presentation will contain forward-looking statements, including, and without limitation, statements about the company's plans, strategies and prospects, including capital structure and go-to-market strategy, the company's future operating results and financial position, the company's market growth and the company's objectives for future operations. These statements may include words such as believe, may, will, estimate, continue, anticipate, intend, expect, could, would, project, plan, intentionally, preliminary, likely and similar expressions and are intended to identify forward-looking statements. These forward-looking statements are based on information available to the company as of the date of this presentation and are based on management's current views and assumptions. These forward-looking statements are conditioned upon and also involve a number of known and unknown risks, uncertainties and other factors that could cause actual results, performance or events to differ materially from those anticipated by these forward-looking statements. Such risks and uncertainties and other factors may be beyond the company's control and may pose a risk of the company's operating and financial conditions. Such risks and uncertainties may include, but are not limited to, changes in market demand, the effects of competitive markets in which they compete, changes in technology, market acceptance of new products, the company's ability to implement its strategies and plans, the company's ability to successfully qualify and sell its products and services and increasing volumes on a cost-effective basis, the company's ability to generate sufficient cash flow from operations to meet its liquidity requirements, the continued impact of COVID-19 on the company's business and evolving legal, regulatory and administrative climate in the international markets where the company operates. Information concerning risks, uncertainties and other factors that could cause results to differ materially from the expectations described in this presentation is contained in the Risk Factors section in the company's annual report on Form 10-K and the quarterly report on Form 10-Q filed with the U.S. SEC on June 24 and August, respectively, both of which incorporated into the documents I referenced and other documents filed with or furnished to the SEC. These forward-looking statements should not be relied on upon as representing the company's views as of any subsequent date, and the company undertakes no obligation to update forward-looking statements to reflect events or circumstances after the date they were made. The use of non-GAAP financial information. In this presentation, the company will present non-GAAP measures of adjusted operating expenses, adjusted EBITDA and adjusted EPS, which are adjusted from results based on GAAP. These non-GAAP financial measures are provided to enhance users' overall understanding of the company's current financial performance and the company's prospects for the future and are not comprehensive of the company's financial results. Such measures should not be viewed as a substitute for the company's financial statements prepared in accordance with GAAP. You can find reconciliations of these metrics to the reported GAAP results in the reconciliation table provided in the appendix of this presentation. A reconciliation of non-GAAP measures to corresponding GAAP measures on a forward-looking basis is not available due to the high variability and low visibility with respect to the charges, which are excluded from these non-GAAP measures. With all that said, I'd now like to turn the event over to Jamie Lerner. Jamie, the floor is yours.

Thanks, Rob. Good morning, good afternoon, good evening to all our analysts and investors from around the world. I'm Jamie Lerner, I'm the Chief Executive here at Quantum. Many of you may recall, 2 years ago, Quantum was in somewhat of a perilous state. And our Board, our executive staff and our employees executed a series of steps designed to stabilize the company. And over the last 2 years, I think our customers feel that they're in good hands. We've relisted on the NASDAQ, our innovation engine is running, we're releasing products probably faster than we ever have before. We've got a strategy that's resonating in the market. And I think we have stabilized the business. And as the stabilization phase is coming to a close, we're moving into a new phase of transformation. Today, we're going to lay out a multiyear strategy, where we're going to move from what was predominantly a hardware company, and we're going to transform ourselves into a software-as-a-service and more cloud-oriented company. Now I'm going to walk through the details and the anatomy of this transformation. And many of you are going to probably find it to be ambitious, maybe even somewhat audacious. So I've asked some of our customers, our general managers, our sales leaders to join me to explain to you not only our strategy, but how we're going to execute on that strategy. The work that we're going to do, what we're going to deliver as we go on this journey to transform from a hardware business into a software subscription and an as a service business. Also we're going to have Fred Moore join us to talk about some of the major industry transformations that are happening that we're aligning to, and we're going to be taking advantage of as we transform the company yet again. And as all this comes together, Mike is going to conclude with a financial model that if we're able to make this transformation. And as we execute the transformation, you're going to see the earnings power of the company deeply transformed our margin profile, moving from episodic sales to recurring and more predictable sales and ultimately profoundly changing the earnings power of the company. So with that, let's jump in. Eric? So when I joined the company and in almost every week that I'm with the company, I spent the majority of my time with our customers. And when I work with them, what I often talk about is, it's not only what they're doing, but why they work with us? What is it that Quantum does uniquely? What is it that we do better than anyone else? Why do you work with us when you have so many other options or ways to store data? And what I hear over and over from our customers is that it's our expertise and intellectual property in unstructured data. And in video and images and things other than letters and numbers that feed -- fit deeply into columns and rows and databases. We're the guys that deal with videos of radar and LiDAR and in imagery and models of our planet, of oilfields. And we've had a huge amount of success in this area in building intellectual property in this area, and we've designed our whole strategy around going after this type of data. Next slide. Now we know that in the next several years, even today, we know that unstructured data is a majority of all the data on our planet and by 2025, unstructured data will be 80% or more of all the data stored digitally by mankind. So we should probably take a minute and talk about what is unstructured data. So to start by defining what structured data is. So if you've ever bought anything online, you probably typed in structured data. You type in your name, first name, last name, fit neatly into boxes. You enter your ZIP code and your address, your telephone number, again, fitting neatly into boxes, fitting neatly into structures that are stored in rows and columns. And that data is mostly letters and numbers. And we don't -- we're not really focused on that kind of data. We deal with data that is very, very different. We deal with movies. They don't fit neatly into anything, right? Hundreds of scenes and music, and the data is not just a few letters and numbers like in your name, but it's billions of letters and numbers to produce a television show or to model an oil field or an oil reserve underneath Earth's crust. Those are enormous models, modeling a nuclear shockwave. Think about hundreds of hours of video surveillance. That's the kind of data that we deal with. And I would tell you that the way humans think about that data is totally different, and that informs our whole strategy. Again, Eric, you can back up the -- what I mean is, if you think in your own life, all of us are surrounded by letters and numbers. We received our cable television bill, our bank statement. And those are letters and numbers that are in those. And I would say most of us don't feel we need to keep those very long. I don't keep my bank statements from when I was in high school. I just don't think they're valuable any longer. I don't keep my cable television bill from 30 years ago. Right? We keep those letters and numbers for maybe a year, 2 years. And eventually, they just become valueless over time, but not unstructured data. I keep my photographs or the videos of my childhood, and I want to pass them on to my kids, and I hope they pass them on to their kids. We expect to keep that data, photographs, videos, images forever. When we go to museums, we look at art that is 500 years old, 800 years old. Well, don't you think movies will be kept for hundreds of years, I think they will. I think Star Wars and other kind of major works, which are effectively works of art, digital works of art will be kept forever for hundreds of years. And -- but letters and numbers, I think companies keep them for 6 or 7 years and then discard them as soon as regulatory allows to get rid of them. And so this data is different. It's handled differently as we think about it differently. Next slide. The other thing that I would share is not only is unstructured data enormous. But it has moved around tremendously. If you think about a movie, it's moved all over the world. It's moved into the home, billions of homes. Television shows spread to thousands of homes, tens of thousands of millions of homes. When a piece of unstructured data is being analyzed, scientists around the world may be looking at it. But again, structured data, for example, if you were going to buy something online, it sits in one place. It sits in one database. It may be backed up somewhere, but it's is in one place. But unstructured data is transported all around the world. And as I talked about, it's often kept for not just 6 or 7 years for regulatory compliance, but it's kept for arguably 100 years or longer. And the other thing that is informing our strategy is that 40 years ago, I think most companies' most valuable assets were physical. An oil and gas company would have listed their refineries as their most valuable assets or they're offshore drilling platforms. But I think if we fast forward to today, most companies and many companies are transforming to where their most valuable assets are digital. I think Disney or Warner Brothers would say their most valuable assets is their catalog that is stored digitally. I think major oil exploration companies would say it's their high fidelity models of the Earth's crust and beneath the crust and the oil reserves beneath the crust are the most valuable asset they have. And more and more companies, their digital assets that need to be kept for decades are their most valuable asset. And most of those digital assets they have, I would view them as mostly unmined, unexploited. If you think about the world's biggest retail companies, they have thousands of hours of shopping behavior on their video surveillance. Who comes in their stores, where do they walk? What do they buy? What are the demographics of the buyer? There's incredible information in that data that is unmined. There are millions of hours of satellite imagery, movie information, news channels, where there is data that can be mined from that, that today is mostly -- has been unanalyzed. And I think that data -- there's incredible value in that data that's going to be reanalyzed and reanalyzed for decades to come. Just think about a human genome, how many times as we learn more about science, as we learn more about analyzing DNA, how genes that may be 30, 40 years old, we analyzed and reanalyzed and kept for hundreds of years as we study disease more and more. Next slide. And so as we think about this data, one of the biggest differences is it's life. Again, an online transaction, you type in your data, and it just sits there for 6, 7 years, and that's purged. The data that we're talking about is totally different. It moves. And most importantly, it starts its life, not on the cloud. Not in the data center. It starts its life out in the wild, right? When we work with a movie company, that data starts on a movie set or out on location. It might start at a sporting event. When we build autonomous vehicles, we have petabytes of data being created on the roads. When we worked with NASA, and other space exploration companies, the data started in satellites that are orbiting our planet. Data comes from airplanes and military equipment. It's just not in the data center. So it started on the edge of the network. Right? So this is not a problem that cloud computing fixes, right? This data is out in our world. And it needs to be captured and then transported into an area where it can be quickly modified, analyzed and insights can be drawn from them. So for example, a movie may be evident, a genome may be analyzed, an autonomous vehicle may be tested to see if it performed properly. Images of our planet may be studied to look at long-term effects on coral reefs or a large force. So the data is moved from the edge into the core. It's analyzed. It's looked at -- and then it's set down, it may not be used for a period of time. So it may be archived. But that archive is alive, it's active because a genome or an old movie or an old sports program or old scientific data, maybe brought back up again but reanalyzed, set down again. So this data is living for not just 7 years, not just 70 years, it may be living for hundreds of years being reanalyzed restudied, retransported around to different people and organizations. And this is the problem that our customers say, we're best suited to solve. And this is where we're going to be focusing all of our energy over the next 5 years. And so what we're building starts to look like a layered cake. And you'll notice our bottom layer is somewhat grey and dull, and that's on purpose because we're deemphasizing hardware. We're moving our architectures that they can run on any hardware. None of our flavor, none of our special I.T. are -- sizzle on our steak is not on the hardware. It's in the software tiers. And so what we've been doing over the last 2 years is separating our intellectual property and our software away from hardware and using commodity hardware where we can. Using basic hardware that could be delivered by any number of vendors, customers could acquire it themselves. And we're making it so separated from the hardware that it can even run on the cloud. If the customer wants to place on the cloud, that's fine too. And where we're placing our energy is on that blue layer, which is our storage software, the thousands of features that we have that allow people to store a video, move a video, archive a video, take unstructured data like a genome or a model of the Earth's crust, study it, analyze it, transport it, protect it. And then what we're doing is building a new layer above that, that you see in purple, which is how that data is organized and orchestrated. We have customers now that have over 100 billion pieces of unstructured data. We have customers that have tens of millions of human genomes that have several billion photographs of the Earth, that have tens of billions of testing models of chips and cars and different technology. They need a model to organize that. How do you catalog that? We have -- the world's biggest news networks have 40 years of their news. They have to catalog that, organize that, search it, place it in different places, place it on the cloud, place it in archive, bring it forward for that night's news. They need tooling to be able to organize these enormous archives, and we are increasingly moving up to that higher software layer, and all of that software runs from the cloud. It's not software that we're going to sell to anyone on a CD or they're going to download it. It literally is where you swipe a credit card and start running that directly off the cloud. And if you look at the bar on the right, all of these capabilities are going to be basically quantum as a service. We're going to deliver them like dial tone. You pay a bill and we give you dial tone. If you need to store videos, we'll just turn it on for you. If you need to move videos, analyze videos, deal with genomes. We are going to just provide that on tap versus the way we are still today, putting hardware on pallets and shipping them to people's floating docks. We're going to be deemphasizing that and moving to the software and as-a-service models. And so if you look at what we've accomplished to date, we came into the business with a long secular decline and much longer than this for almost 10 years, the company had been in decline. And when we first started talking to many of you, we made an admission that many of our older businesses are in declines that we can't stop, just older technology that's being used less and less, and it's in a long decline. And the company hadn't built new products in quite some time. So there wasn't an easy way to stop that. But if you look at the top blue line, we stopped the declines. And we did that by introducing new products in new markets, in growing markets and the growth of our new products counterbalance the declines and we stopped at declines. COVID-19 hit, we did have some further declines, but many of you can see we're climbing back out of that again. As the pandemic subsides, we'll get back up to historic levels and well beyond those historic levels. But the other thing we did during this period is, look, what we did to our margins. We completely changed the direction of our margins and begun to get margin expansion. As we built more about -- our newer products are more valuable in nature, going after solving more complicated problems, and we're increasingly solving them through the differentiation of our software or getting better value for that business model. And while introducing all these new products and transforming our business, we tightened our belt. We conducted these transformations on a budget, and you can see with the lowest line that we brought our OpEx down while we made these transformations. And we're going to continue these efforts in an accelerated way as we step into our transformation. Next slide. So the big moves that we're making is we're transitioning from hardware to software-defined architectures. Then this software can be run on the customer's premise on their own hardware, it can be run on the cloud. It can be run by third parties. And increasingly, this software will be run at some form of either a as-a-service offer or managed service offer. These services were building service delivery platforms where we can deliver these services at scale in a very automated way so that we can scale our business without piling in many, many new hires we're going to use technology to scale versus large-scale recruiting. And we're going to be entering the data management and orchestration space to not just store our customers data, but help them catalog it, organize it, gain insights from it, move it to where it needs to be. And simultaneously, we're taking this technology outside of our traditional backup and media and entertainment spaces. We're moving into surveillance. We're moving into autonomous we're moving into new geographies. We haven't played much in Asia, even though it's the fastest-growing markets in the world. We're going to be going into these new markets. And we're going to be helping the world's largest company solve the need to archive their digital assets, whether they're scientific in nature, artistic in nature, they're part of our mankind's history. We're going to help them store these digital archives for 100 or more years. And we're moving into these faster-growing, larger markets, and we're going to achieve a very different outcome. Today, our sales are mostly episodic. They're onetime in nature. We're reshaping to recurring revenue. We're going to start talking in terms of total contract value, monthly recurring revenue, annual recurring revenue. As we move to software and away from hardware, we're going to be reshaping our margins drastically, as you can see on this slide. We're going to be disciplined. We're going to make these transformations without dumping a ton of investment on the business. We feel we can generate the money that we need, the resources we need, hire the talent we need to make the transformation with the OpEx envelope that we have today. And if you look at the earnings power of the company, it's just dynamic. We're looking at 3x, 300% EBITDA expansion, EPS of over 6x of how we're going to be reshaping this business. Now I know many of you think this is ambitious. It may sound somewhat audacious. From my point of view, this is a transformation that we have no choice. The world is moving to software, the world is moving to cloud, the world is moving and as a service, and we simply have to make this transformation. And given the last 2 years of accomplishment, I have every confidence that we're going to make it. And what I think is going to surprise you in the next hour is just how much of this we've already done. So I'm going to hand it back to Eric, who's going to show you a customer video, and then we're going to get into our general managers who are going to walk you through exactly how we're going to carry out this transformation.

Thank you, Jamie. My name is Eric Bassier. I lead product marketing here at Quantum. And what better way to bring some of those challenges and our strategy to life than hearing from our customers. So let's hear from one of our long-time customers, Major League Baseball Networks. So Tab, you guys are right in the middle of the season now, what's next for Major League Baseball Networks?

For us, at this point, we are coming up to the post season. The post season like none that we have seen before, it's going to be quite exciting. And for us, the challenge is that we are going to be doing the world series with very few people on-site, remote operations globally, very, very exciting and really impressive when you look at the systems and the partnerships that we have developed and are continuing to develop to ensure that this product continues to grow and be as well received as it has in past years. In our archive environment today, we are averaging over 2,500 archived jobs on our systems. Moving 50 terabytes a day of new data into the archive. In the last 4 weeks, I have had peaks that go up to 5,600 jobs and moving 102 terabytes of new unique content into our archive in one single day. At the same time, I am reading out of this active archive to the point of, on average, 2,000 jobs a day, which is a total of 40 terabytes of content read per day. However, we have seen spikes up to 4,600 jobs and over 100 terabytes of content recalled from this active archive in one single day. And this is a snapshot of just what's going on in the last 30 days at MLB network. For us, forever is going to be hundreds of years, hopefully, more than that. But the key for us is that we have content in our archive may be still images and things of that nature. Film and content that dates back to the early 1900s and images that goes back into the late 1800s. And we need to have all of that stuff at your fingertips. You're going to want to be able to access Derek Jeter's first home run. You're going to want to access Derek Jeter's 3,000th home run. And for us, that Quantum core infrastructure backed by an archive library is critical. We are seeing editors that have in the beginning of the year, said that they had to be in their room with the talent. And today, they are saying, I can do this work at home. I can do this work remotely and I can get a very high-quality product out. They are -- we continue to develop tools. We utilize the cloud to facilitate quality control and review and approval techniques, we are still utilizing a single core with remote access into those machines. So the content, which is the valuable intellectual property, the value of the corporation of MLB network is the archives, is the baseball content. That baseball content is still serviced but is still protected. And available to the editors, the creatives to create and tell their stories, do their highlights. We're a storytelling company, and we're trying to tell the best compelling stories we can about baseball using these tools that facilitate speed and quickness and reliability taking our product to market.

Tab spoke about our StorNext high-speed file system as well as -- as they're moving to a hybrid and multi-cloud infrastructure. To talk more about our solutions in this area, I'd like to introduce Ed Fiore, our General Manager of our primary storage business. Ed?

And it's nice to see you all virtually, I joined Quantum about 5 months ago. I came in through the acquisition of a company called Atavium, which when Jamie was talking, we talked about the visibility to data and how important that is in order to know where it's going to live or where it needs to be, we have a core technology that Atavium was working on. And we're bringing that into the StorNext product families and all of the quantum products. I have a pretty long history in storage and storage networking, companies like Storage Tech and Cisco, in my early days. I was the founder of the Isilon platform team. So I've built that from the ground up in Minnesota. I joined Compellent kind of its mid life. That was acquired by Dell. And at Dell, I ran all the IP assets for Dell storage prior to the EMC acquisition. From there, I was one of the co-founders and the CEO of Atavium that I just mentioned. Next slide, please. Our portfolio is really a rich portfolio. It's about TAM expansion. It's about where the world is going. If you look at what we had traditionally, we had our Quantum StorNext, it's very fast file system for really demanding workloads, genome sequencing, 4K streaming is what it was built for. It also had some really good abilities to place data, which is really where the world is going, right? If data is not going to live in a single data center anymore, the ability to move the data that you need from the cloud or from the on-prem to cloud or from flash to bulk, from tape is really important. The ability to manage that data with -- data that's more than just timing and usage, it's about adhere needs versus time and usage. So StorNext is a great product. It was born software-defined. It is going to be software-defined going forward. We basically will have another side of this, but talk about running in virtual environments as we go forward. Some of the things we're working on as we go software-defined, it's really about a recurring revenue model. So as we become software-defined and less dependent on the hardware and we come more about data under management, data will live -- I'm sorry, it's still on the slide, Eric. Go back 1? Thank you. So we'll become -- we'll really be software-defined, go to a perpetual licensing model dollar per terabyte per month is data under management. We have a VS-Series product line, which is about expanding TAM. We have a pretty broad portfolio here. We go everywhere from a couple of hundred cameras to thousands of cameras. We're leveraging internal IP like our active scale and our StorNext as we build these bigger systems out. We've got several proof points in very large environments like Tokyo Airport. We continue to grow this market. This market is about infrastructure building. So it's a little bit of a different buying cycle. We're not really buying through IT and these aren't IT savvy people. These are people that build buildings for a living. These projects can be very long, timely or take quite a bit to actually build out a building. One of the things these guys are looking for is easy to use, and some of it's just going to support their environment. So as we integrate these products into our CBA or our phone home capabilities, we can monitor these remotely for the customer, we can leverage our service organization in this environment. And it really is something that this market needs. There's a lot of niche players in here that don't have that kind of environment. So this is about TAM expansion for us. And we're doing -- we're actually starting to grow that market as we speak. It's been a little bit of a long road, but we're getting good traction now. The R-Series product lines, it's really about edge products. This is about data that's created on the edge, it needs to come back to a data center or the cloud, it's going to always want to run more compute on it. If you're in an autonomous vehicle, you bring that data back, the vehicle itself is doing analytics, and it's got compute, and it's doing jobs, but you're going to bring that data back, and they're going to continually analyze that. Every time they'll do a software update, they'll continually analyze it again. So the R-Series for us is about the portability of data. If you look at markets like IOT, onset movie production, theme parks, automotive and the hyperscalers, all these teams or groups have the need to basically make data, get data from the edge back into a data center. This market is growing quickly. There's not a lot of competition in this space for these types of environments. We basically have a portfolio now that goes up to over a petabyte in a portable fashion, and it's growing for us very quickly. All right. As Jamie said, we're really focusing on software-defined. If you look at our legacy architecture today, the picture on the left is what you would basically deploy as of today, very quickly by the end of this year, we have a new set of products coming out and really a new software stack that really is about running anywhere. It can run in a hypervisor, it can run in the cloud, it can run at your hardware or our hardware. It doesn't matter to us anymore. The value of our product is in the software and the ability to manage the data and protect the data long term. So we're focused on that. Again, this is about margin expansion. There's much more margin in software than hardware. Obviously, hardware is being commoditized. So as we move forward, that's the path we're going down. The other thing is, as we're managing data in the cloud, there's also a lot more data that's moving to the cloud that needs to be managed on-prem. There's many customers out there that still want visibility to their data in the cloud and still want to manage their data in the cloud, even though they're on-prem, we can charge for that data under management, which is really a new thing for the market. It's relatively new last couple of years. The other thing we run in the cloud. So as StorNext has been software-defined from the beginning, its ability to run in the cloud has been pretty -- was pretty lightweight lift for us. We've got a couple of customers that are running in multiple clouds now. We'll see that product be productized this year. And it's no longer about hardware. It really is just about software and data management. A lot of things we're doing are still focused on ease of use. If you look at certainly the M&E market, these people -- that market is not IT savvy and it's lower end. And so giving people lower end products that are just easy to use, plug and install, click and play is really what we're focused on. So that's kind of the future of StorNext and our software-defined. Next. So cloud, if you look at where we are in cloud today, we've had a couple of proof points. We've been turning to the cloud since 2018. This year, we actually gave customers the ability to see their data and manage their data independent of the storage, the StorNext being in the cloud, so native object used. We're running in the cloud today at a couple of different customer sites, and we're continuing to make sure that we have the use cases down for that. COVID accelerated a lot of running in the cloud. If you can imagine all the individual people that now need to work from home, they want to be able to get those data sets -- to those data sets, so they're pushing them to the cloud. If you look at the long-term and the vision here, the reality is, the data that needs to be in the cloud needs to be understood better. There's still a lack of awareness of what is all the data that I own and where does it live. And so data management and orchestration of that data -- not just data -- not just curing, but true, where does it need to live? What cloud does it need to live, when does it need to live there? Is what we're working on. All of these things are software. They're all margin. There is no hardware to deploy in a lot of these cases. As we live in the cloud, we'll make it very easy to spin up and spin down in the cloud. People want to have a file system that is independent of the cloud render's file system. They want to be able to move from cloud to cloud. So in order to actually do that, they want to still have control of their software and their file systems that they're running. So they want that same ease of use or same workflow to go from AWS to Google to Azure. All right. Next slide. So where we're going. If you look at the market today, storage has really changed. It's no longer about just storing the bits, it really is about trying to understand the problem the customers have. And if you really go in, talk to customers, it's pretty funny. They all have good stories. You can't believe the number of customers that you walk into, and they'll say, I've got a petabyte of data. I have no idea what it is. I have no idea what to do with it. And they're fearful of it. They don't know -- they can't delete anything. They're not sure if the Vice President's going to come and yell at me. They don't know who created it. They don't know how important it is to the business. And that's really getting to be the challenge for these customers is being able to understand their data. If you look at a legacy storage system today, if you wanted to search a petabyte or 100 million files or a couple of hundred million files, it could take days on legacy systems. And it's a challenge for them. They right custom scripts. The guy who wrote the scripts who leaves the business. And they start these runs on a Thursday and by the next Wednesday, they may or may not be done. By the time they're done, the data they have is old, they're looking at their data at siloed, and now they're basically worried that the cloud is a new siloed. So those are challenges for them. If you look at how they're using the clouds and archives, they just continually push. They don't know what it is. They don't know who created it. They don't know what value it has for their business. So they're really just creating big mountains of data that they're really struggling with to understand. Next slide. So what are they looking for, right? They're looking for classification. Hey, I don't want to have any impact on my workflow, I want to be able to classify my data. I want to know who create it, I want to know what they created it for, I want to know what movie it was, what genome it was, what project it was. They're looking for that kind of intelligence of, hey, give me a breakdown from a business knowledge, what is this data that was created. On the searching side, they want to be able to search instantly. They don't want to wait a week a day a month. They want to know, hey, I want to know who just created 100 terabytes of data this weekend and blew up my system. We have a university here in Minnesota, where one of the scientists came in, in the morning, and there was 100 terabytes created over the weekend, basically blew them up out of storage you can literally look at interface and say, oh, I know who created this and you top it down the shoulders and just get rid of it. So those kind of real-time search and analytics are very important. And they're very -- there's very few storage systems that can do that. And what we're building today does. Data placement, it's no longer -- if you look at traditional storage, data is about movement over time. No one uses data for the last week, I'm going to move it here. No one uses data for the last month, I'm going to move it there. That's not how data is used. You heard Major League Baseball made the comment, hey, if I need to get a scene from Derek Jeter from 2012, I need it today. I don't want it tomorrow. So it's not about data. It's no longer about data -- movement, it's placement. We need to be able to go get data. We need to get it when we need it. One of my favorite stories is the DreamWorks guys. When I'm working on Shrek 84 and you need a scene out of Shrek 3, I need it tomorrow. I'm going to get it right now, I can't wait until tomorrow. So that's still a challenge that people are working on. We talk about the cloud. And today, the cloud is really about moving data. It needs to be seamless. You will see that the cloud needs to be leveraged for orchestration. It needs to be leveraged for compute. It needs to be leveraged for quite a few things. So using the cloud as kind of an infrastructure, part of a bridge, which is why we use the term bridging, is really important to every customer. Every customer we talked to have some use for the cloud, whether they're going to run their applications in the cloud, whether they want to use it for elasticity, it doesn't matter, they all have it. And then the ability to delete and archive things is just a challenge to them. They don't know what it is. They don't know how to archive it. And these are things that you can basically go to any customer and they'll find the same challenges. Next slide. So how are we solving this? So this is a technology slide that we built from Atavium. This technology is now in Quantum. We're spending time iterating this with all of the Quantum products. You'll see very specific proof points of this very soon, where we actually can manage all the Quantum products with this classification. Fundamentally, on ingest, we add these things called tags. These tags are really classifiers. We can add an additional tags to these. So you can tell where there are. But it's very easy for a user as data is being ingested, it's this movie, it's this user, it's this DNA sequencer, it's this camera. So the ability to identify where the data was created, what it was used for and where does it need to live is all part of classification. Without classification, you really can't do anything else with the data because you're just guessing at that point. Now that my data is classified, the ability to move it. How many copies do I need to do? Do I want to copy on tape? Do I want to copy in an active scale in an object store? Or I want to copy on the cloud? Or multiple copies in the cloud? So the ability to move that data, the ability to give access to that data is equally as important. If you look at the movie industry today or even the scientific community today, they give people access in a direct restructure, who's got access to this directory. Well, they don't want that anymore. They want -- I want to give access to this set of movie scenes or I want to give access to this set of sequences. And they don't have that ability today, but with classification, you can do that. You can really narrow your scope of what you allow people to see and work on. And you can narrow that scope of what you move. So when you're moving projects to the cloud for compute, instead of moving the whole project, you can move just the pieces that you want to move as you go forward. And that's the beauty of classification. All of this is software. The value of this is all -- it's all data under management. It's all recurring software license. This is a dollar per terabyte per month of data under management type charge. So it really is going to change our trajectory from basically a hardware company to being a software and data under management company.

Really exciting. And thank you very much. Before we turn it over to Bruno, our other GM, we wanted to hear from 1 additional customer. In this case, it's the GWDG, which is the compute center for the Max Planck Institute and part of the university system in Germany, and they've effectively built a private cloud archive with Quantum software and Quantum tape. So let's hear from the GWDG.

My name is Ramin Yahyapour. I'm Professor in computer science at the University of Göttingen. I'm also Managing Director of GWDG. The GWDG is a scientific compute center, serving the Max Planck Research Organization as well as University of Göttingen. The Max Planck Research Organization is a distributed organization with 83 institutes distributed over Germany. It's a premier research organization, so that group that is doing basic research and also famous for winning a lot of noble prizes.

Your center has done and you guys have been a customer of Quantums for many years and a user of the StorNext file system for many years. Can you tell us how you use StorNext today?

We have a very large computing center in Germany. So we have multiple service offerings to our customers from high-performance computing, to data management, data analytics, also archiving. And we are with StorNext for quite a long time very successfully. We use StorNext as 1 of our key strategic architectures for data management. So in total at GWDG, we are currently around 50 petabytes of data and more than 30 petabytes of data are managed through StorNext. We have around 3 billion file objects currently.

When it comes to the data that's stored in the archive for the research scientists, how long do you need to store that data and protect it?

That's a good question. It's heavily debated in science how long we should store data. In Germany, we have currently something called scientifical practice, which mandates us a lot of our funding agencies to store research data for at least 10 years. So that's the minimum time. We have a lot of customers who ask us to store data for much longer. For instance, we are also the storage back end for several libraries here in Germany, for instance, digital and national library stores their data with us. And for them, of course, they would like to have the data archive forever. So between 10 years and an ever. Besides hierarchical storage management, StorNext is very useful for us because it visualizes the physical storage of back end. So it allows us to have different types of storage from different vendors and basically channeling it through StorNext to our customers. Our customers create data, they work with data, and it's quite well-known that scientists are not very eager to do active manual data management by themselves. So thinking about when to move what kind of data to a certain place and [indiscernible] it and so on. And therefore, we need a solutions that help them and having file system as an interface, and have the ability to automatically apply policies to migrate data to tape and start archiving this, it's a very easy to understand interface for a lot of our customers. So they have connected file system, which they work, do their research. It's also quite well-known from statistics that a lot of scientifically generated data after a certain time are not accessed very often anymore. So there is a time frame of a few months where data is very interesting. Then it’s put away. And then the market situation, that part of the data becomes interesting again. And HSM through StorNext is very nice to basically cater to that kind of use case. So that the data is still visible to all of our users. They can't see whether it's on tape or on more expensive disks or even more expensive SSD devices. And by just reading files again, it gets automatically read from tape to files system, but that's not seen by the end users. And so it's very nicely fits to the research use case we have. StorNext is a strategic component in our overall architecture. We use our next as our main offerings for file systems and data management. That's the reason why the majority of all data that we have is managed through StorNext. It has 2 main benefits. One is it virtualizes abstracts from the actual storage solution. So we have the flexibility to have different solutions behind StorNext and also migrate between those solutions whenever necessary. And we use very heavily the heretical storage management feature to able to migrate data depending on excess time, some kind of policy from tape to disk, from disk to tape, however we need it to.

So another very interesting use case talking about large-scale archives. And to talk more about our strategy and our solution set for large-scale archives, I'd like to introduce the General Manager of our secondary storage business, Bruno Hald. Bruno?

Thanks, Eric. Hello, everybody, and welcome to the Quantum Analyst and Investor Day. My name is Bruno Hald. I have been with a Quantum for many years, have been launching a lot of products in the -- mainly in the data protection, archiving and backup space for ADIC and Quantum mobile, as I said, in many, many years. I'd like to today talk a little bit about what we're seeing in the market. There are very interesting changes happening out in the market and how we are reacting with our products to these changes that we're seeing out there. So next slide, please, Eric. So this is the product portfolio. We're looking here that we basically classify as secondary storage. Secondary storage being more the data lake, if you will, products that provide the capacity to store massive amounts of data that are being generated. We have heard this from the 2 customer video that there's many companies, enterprises, institutions that generate a massive amount of data on a daily basis. And that data needs to be hosted somewhere else. Even the cloud providers, I mean, if you think of an S3, for example, this is an object storage, right? So the products are showing here, left-hand side is our object storage software. It's ActiveScale product, which was recently acquired from WD, but we have been reselling this product for many, many years. So we have plenty of experience with this product itself. Then we have our tape storage systems. So these are systems that are more used for really very long term, highly efficient and cost-effective storage of data. And I'll talk about this a little bit more in the next slides. This is the foundation of a lot of the lower cost tiers in the cloud. And we have both a lot with the hyperscale providers in the past. I'm going to talk about this a little bit going forward. And then we still have our DXi series. This is our data deduplication appliances right now, but we're also migrating this into a software model. But we're probably staying also with appliances in parallel for the time being. This is needed for more your legacy market storage traditional backup, but it can start moving into other areas of protection as well. And then, of course, all these products are being serviced with our magnitude of service offerings that we have. Customer manageable all the way to fully Quantum operated and managed, if you will, right? Eric, next slide. I want to talk a little bit about what we have experienced in the tape market over the last 2 years and our experience with the hyperscalers .If you look at the development of the overall tape market, your classic IT market, it's definitely shrinking over the years, right? We're seeing that data is being migrated to the cloud. But specifically for tape, there was an interesting inflection about 4 to 5 years ago, we started working with one of the biggest hyperscalers in the world to see if they can make tape work in their use case, in their environment to provide a layer of storage that is extremely cost-effective and can be targeted towards the archival space where you can think of it like the data that is rarely accessed, right? So that just create this big, big data lake and be able to offer this at a very low competitive cost to the customer. So about 45 years ago, we engaged with this hyperscaler and it took us about 2 years to get this in collaboration with hyperscaler to a level that it was production-ready, that it was worthwhile in terms of scale that they were looking at. And then about 2 years ago, the decision really was very clear that this is something that works for them, and they basically work into production. What you see on the right-hand side are a few statistics around what we have going now in the hyperscale market. We think we're #1 in this market because we work only in the development there. We helped actually to get credibility for tape into this space. There were other hyperscalers that have tried tape before that actually gave up on it. But we are at the point now that this is absolutely the best low-cost solution that can be offered there. What we have deployed is about 20-plus exabytes of capacity. And if you lay these robot systems out, you would have about 3.7 miles of hardware lined up, all tape basically in the cloud right. And then additionally, we're expanding the reach into more hyperscalers. We're talking to a number of the very large hyperscalers since -- while they are in different stages of trials. But very clearly, the pressures on everybody. It has been proven, as I said, that tape works well in this environment now and they're moving forward with tape. Eric, next slide, please. So just a little bit detail around what these deployments look like. This is our scaler Ti 6000. This is one of the options we offer in the hyperscaler market. Think of this as 1 consecutive product. So the top picture you see there is literally 1 product. That's the way it gets, right? These things get molded together, the robots move from left to right, horizontally. They're in the system. There's about 190 petabytes in a system like this with dual robot system. It's highly reliable. This is your legacy enterprise IT high end kind of equipment that we have moved into hyperscale usage as well. However, what we're learning and what have been learning over the years now is that this is a little bit an odd kind of thing for the hyperscalers. It doesn't fit necessarily the standard deployment model in the data centers, and we show what their standard looks like. It's 19-inch rack or something similar, right? A smaller type of building block that's easier to deploy for them. And that's really what the hyperscalers are looking for, right? So when we advance to the next slide, Eric, I can talk about what goes into a decision at the hyperscalers. Overall, what products we use for what level. I mean it's all about TCO. It's all about total cost of ownership. What can I offer the service for dollars per terabyte to my customers and therefore, what can I spend basically on everything that includes to build that solution, right? At the hyperscaler. So it starts certainly with the expense for the equipment itself but there's much more around this, right? When you look at tape today, and that's the picture you're looking at here, it is still the overall lowest cost storage medium in the world. So therefore, the hyperscalers have to look at this. With the massive amount of data that customers are asking to be installed in this space, they need to have the lowest cost there. And that's what tapes serves today. However, besides just equipment, as I said, there's more consideration, right? Yes, Eric go ahead. So for your spaces of consideration, clearly, how much of this capacity can you basically pack into a certain volume of space, if you will. And we have a product out in a market. It's our rack-based modular system. This is the highest density system in the market right now and very attractive for this hyperscale users well, right? It competes very well with the more continuous build systems like I showed with Ti 6000. Then the other thing to consider is power and cooling. Very clearly, tape is absolutely the lowest power consumption media that's available right now. However, tape does come with a few challenges, especially around environmental. So you got to treat tape well, you got to keep it happy in regards to temperature and humidity so that you can really guarantee your integrity of your data on the tape. So there's a little bit of investment that's needed to make sure that the environment is good for where you operate tape in. But this balance is well out in regards to your classic power consumption. Hard drive people, they are definitely working on technologies also to get the power down. But at this point, tape is absolutely lowest power down. But at this point, tape is absolutely lowest power. If you look at the tape cartridge itself, it doesn't consume any power. It just sits on the shelf. It only consumes power when you put it into the drive, and you need to read, write, right? This product is designed for more inactive type of data. So it's a good assumption that you can expect that the tapes sit on the shelves a lot and don't consume a lot of energy, right? And then the other big component that goes into the TCO conversation is really how easy can I manage service, install these systems, you have to think about scale, you have to think about massive scale, and you need to design your systems for this. It's not -- I'm not just dealing with 1 or 2 systems, I'm dealing with hundreds or thousands of these, right? And that's the way we got to really design and think about when we produce these systems that it needs to be easy to handle. So we'll show you here an example for a typical crew, a customer replaceable unit. It's easy to basically exchange failed components for their operators. Their data center technicians are not highly trained on these individual devices anymore. They're looking for generalists. And therefore, the equipment needs to be really designed the way so that it can super easily be identify what's failed, how do I replace it, and then without much reading, whatever, they just walk up, replace the component, stick the new one in, and they are ready to go. So that's another really big component for the TCO conversation for the hyperscalers. Next slide, please, Eric. So as we have worked with the hyperscalers, and I mentioned the -- let's say, a little bit construction project around Ti 6000, it's a little harder to deploy, special spacing required. We have pushed out a path now with an architecture that we call Rail. We've done an area of independent libraries. This is basically a concept where we're now providing product that fits much, much better into the hyperscale data center deployment model. It's a rack-based system. So think of it a robot going up and down this rack, it's individual systems, and then this customer basically arrange a code across a series of these systems. A typical cluster that basically goes into these data centers is comprised out of 5, 8 or 10 of these individual racks. And the data is spread across these individual racks. Think of it very similar to your hard disk rate systems, right? What you get with this kind of architecture is best in storage density, it is super easy to deploy, as I said, you roll this in, you set it up within 1 hour, you have 1 rack up and going, that's 15 petabytes of storage ready to go within an hour, really, really quick. And it's modular, it fits with system and you get higher performance because instead of 2 robots in 1 system, you now have basically 1 robot per rack. If you have 10 racks, that's 10 robots can operate at the same time in parallel. Right. So this is an architectural -- we are in parallel offering to our hyperscalers. There's a choice to be made on their side, some still need towards the enterprise-class systems, but this is also a very interesting concept for hyperscalers, and we get a lot of interest in this model as well. Eric, next slide, please. So and then we have -- this came up now. So we have seen this in the videos. Ed talked about this, and we will hear from Fred in the next presentation as well. Customers are looking for the storage industry to solve this 100-year data life cycle issue they have. Data doesn't get deleted anymore. Data needs to be available. Data needs to be securely stored. A lot of data that's created, a lot of data needs to be searchable and need to be available all the time. So this is a challenge Quantum is going to embrace. And there's -- clearly, you have to think about it, what the challenges are, right? You got to be able to store this data for 100 years, but we also be able to read it back in 100 years from now. You have to deal with generational transitions whatever certain technology goes end-of-life and you have to transition over to a newer technology, right? So all these things need to be considered, but we're creating a framework that really supports this. And the key here is really the software-defined approach to this so that you basically have your object storage, software, in this case, our ActiveScale software, for example, is the intelligent and middle with data classification on top of it that Ed talked about it, that's really a combination of all of our products. And at this point, really, your software is really the critical piece. It doesn't matter too much anymore what hardware's underneath it. And you can swap out these components the way these systems are now designed. And you can see this with RAIL as well. It really helps nicely to update generations. You can swap one rack out and bring new hardware and migrate data over easily, right? This is a product that we're offering and we're working on. End of the year, we will have a, I think, a really exciting offering in the market that we're targeting for enterprises that want to have on-premise storage. So it could be a combination with cloud. For various reasons, there is still a lot of interest out there in terms of having your storage on site, it could be for access speed, it could be for cost, it could be for just we don't want our data in the cloud. So there's many reasons. This is a product that we're super excited about. And as I said, we're going to launch this end of the year, early next year. This is basically your cloud on-premise. It will be fully deployed by Quantum. It's an as-a-service offering. So you're not buying this product. You're basically -- it's a subscription-based implementation, right? So we're taking care of this end-to-end. We're servicing it. We're maintaining it. It's hooked up to our cloud diagnostics. So we have full insight on how well it operates. And it can scale out to massive scale. And again, this is a product that is really interesting for all areas of industries, right? Entertainment, research institutes, surveillance, I mean any really industry you can think of has the need for this type of product. So we're super excited to close this out. Overall, we see a lot of interest in the market. Data keeps growing. 60% of all the data is generated will land basically on these secondary storage layers at some point. So there's a massive amount of data that's out there that needs to be treated and stored for many, many years. And we're looking forward to meet the challenge out there. Thank you very much.

Thank you, Bruno. Thank you, Jamie. Ed and Bruno. Jamie laid out a pretty audacious vision for where we're going. And I hope what you've seen so far is much of the engineering work has either been done or is underway. StorNext is running in the cloud today with some of our customers. And we've built expertise and a market leadership position in very large-scale archives. In short, we're kind of starting to pull away from the competition across the board. To give an industry perspective on kind of the future of data and the need for these 100-year archives. I'd really like to welcome Fred Moore. Fred Moore is the President of Horizon Information Strategies and is probably one of the world's foremost experts on large-scale archives. So Fred, thank you for joining us today, and welcome.

Thank you very much. It's a pleasure to be here, and I wish I could see everybody out there, but thank you for attending this seminar and conference on. Really what we're going to talk about is the archival strategy and challenge ahead of us for the next 100 years. And it's a big deal. We have to prepare for it. I think you've seen through the Quantum speakers this morning, a whole lot of things that lead us after putting this whole picture together and building the foundation for the future. Next slide, please. So our agenda today is going to take a look at this demand curve and what's really behind it. We're going to look at where the data fits by class and storage tier. So as we saw earlier, data classification as a key piece of this. We need to understand that very clearly to begin to manage what is going to be a zettabyte era here that lies right ahead of us. Now we're going to talk about also why is archiving so relevant? We think of archived data kind of like old furniture, you put it in the attic, you don't use it again, but that is not the case with archives anymore. They become alive and they've got tremendous value that we are beginning to realize. We'll look at what applications are behind the avalanche of archives. I want to talk about unstructured data and structured data both and that shift that's underway. We're going to look them closing out at what a 100-year archive looks like and we'll talk about the hyperscale issues that are absolutely paramount as we go forward in this model. So we begin by looking at probably the best projections that we have in the industry to look and I'm going to go out 5 years. We talked about a 5-year view on a lot of things, but we will even go further than that later on in this conversation. But if we go out to that 2025, we're looking at about, the best estimate, 175 zettabytes of data created. The anomaly here is that a lot of that data is not stored. It's very transient. It's started at the end of the session. When you log off of your computer, all the temporary files go away, everything that was generated to support that activity or that session vanish and the data is not permanently stored. And out of that, we see about 95% of the data that is created really never does get stored. So that leaves us in 2025 with the best estimate of around 7.5 zettabytes of stored data. So I think this number 175 has been thrown all around the world as what the storage challenge will be, but the reality is we'll never see most of that data more than a few minutes during the transaction. So we take a look at that data, that 7.5 zettabytes in the little globe on the left, and you'll see 175 zettabytes created 7.5 zetabytes stored and anywhere from 60% to 80% of that data by 2025, will be, what we know as archival data. That is data that's not used that often. And when it is used, it has usually to be used and requested pretty quickly. But it is the largest storage category, largest storage class of all data. So today in 2020, best estimates are 60% of all data is archival. It's growing faster than any other segment. Right now, we have a situation with the pandemic going on that we're pilling up data faster than we can analyze it. Therefore, it becomes archival on first instance of creation, and it sits there, maybe we don't know how long, years may be before somebody can get at it and deal with that. And we don't know the value of this data. We don't know the value of the pandemic data, all the test cases on vaccines, all the demographics about who has it, what age group, what their demographic situation is and their prior conditions and make that correlation analysis and modeling to go with it. That's all to be done against the future archived challenge. So if we look at our world today, data is the new IP. It's the most valuable asset for almost every business in the modern world. And we're going to have to really have a tremendous boost here from software to ever do much with this, beyond what we're doing now, which is basic storage and retrieval capabilities. So we're going to look at deep archive deep machine learning, artificial intelligence to help us take that data and then map it to the optimal storage class in tier. So in today's storage industry, there are 4 de facto standard tiers in the storage infrastructure. The top-tier is the fastest. That's all the nonvolatile memory technologies. You know that as flash memory, DRAM and those type of things, but mainly it's flash memory today used as storage subsystems. And about 10% of all the world's data today, if uniquely mapped, would fall into that category. Performance data, this is high-performance disk, enterprise disk. This is where most of the databases are. That's about another 10% of the data. Mission-critical data sits on this very often. The active archive tier are large capacity disks that don't scale up very well on performance. They do scale on capacity. They're becoming the active archive layer and also the layer for lower activity files that are online that aren't used that much. But the focus here is really on the bottom deep archived tier, where about 60% are roughly today in 2020, about 4.5 exabytes -- or zetabytes of data actually reside on that. And then finally, people often ask me, they'll say, Fred, you know, where does the cloud fit in your pyramid? You know what Tier is that? Cloud is not a tier, it's actually a service. The service -- the cloud services use all of these technologies above. They use solid state, they use rotating disk and tape, and they have services priced based on where your data sits on these various technologies. And then finally, at the bottom, we have what we call [ air gap ] technologies. These are off-line storage vaults, for example, they can be in a tunnel, on the mountain somewhere. They can be transported to an offsite location by truck and this offline removable world has renewed meaning today given all the natural disasters and threats we have to the electrical input to a data center. When electricity goes out, you need another way to move data and that's on removable technology. And the fastest way to do that actually is to put it back on a truck and get it out of there before the flood, the hurricane, the fire or the terrorist attack actually hits. So that's what it looks like if optimally allocated. So why is archiving becoming so relevant? Well, by 2025, if we skip ahead about 5 years, we're going to look at about that 7.5 zettabyte number to be stored, 60%, maybe as much as 80% will become archival if it keeps piling up faster than we can analyze it. And that's growing dead on the 25% to 30% a year net growth compounded annually. I mentioned it's accumulating faster than it's being analyzed. So therefore, it's instantly archival until somebody can go back and touch it. For most data, though, that's created -- it's created an initially stored on disk, and it will sit there for 90, maybe 120 days before it reaches archival status. But at that point in time, it is not economical to leave data that is never being used to sit on spinning online technology, consuming energy all the time. That's a bad economic model, the TCO numbers show that always to be a big advantage to move inactive data off of active technologies. And the other thing we're seeing right now is archival data is seldom deleted. People just don't have the time to resource or the intelligence to go back in and do that. So as I think you've seen through several -- we talked earlier this morning, archiving for 100 years is now common, maybe forever. I mean, will we ever delete the Super Bowl digital video content from any Super Bowl? You will want to see those for thousands of years, maybe ahead and see what that was all like. We've gotten the vast amounts of untapped data. We don't know what is in there, but people are now getting very curious what was all that data I captured. And this is the great challenge here is to unlock the potential of the archive. So making the archives accessible is potentially the CIO's biggest challenge in the storage space as we go forward. So you wonder what applications are fueling this archival Avalanche, I've listed several here, there are several more. But you can take a look at what the application is and a description about what -- about that application drive so much archival storage. We're seeing it throughout the financial world, health care, high-performance computing, media and entertainment, I think that's been covered pretty well this morning already, a tremendous amount of archives there because the M&E people go back into archives to recreate new movies based on things that they've already got captured. That's certainly an issue there. And physical security and surveillance becoming a bigger issue every day and particularly as surveillance retention periods are increasing. We saw the mlb.com video, which is a primary example of how to archived data and tag it and create metadata around all of that streaming video for Major League Baseball games to make it accessible very quickly when it's needed on demand. So if you want to see, again, Derek Jeter's 3000th hit and you want to get that data by the end of the day-to-day, you need a way to find it without reading everything in an archive from beginning to the end. Metadata and objects of the package that goes around that, that we'll talk about in a second, are very, very helpful. And finally, of course, everything out on the edge, this is all the mobile technology, we call it the IOT, really the Internet of Things, autonomous vehicles, generate all kinds of data. I actually have a German car service business where I live in Boulder, Colorado, and I don't go look for somebody that can come in and change oil and spark plug anymore. I need somebody who can run Linux and put in a USB into a car, run diagnostics, do some regression testing and figure out exactly what series of unpredictable events caused this particular check engine light to come on. So it's a new mentality in the automobile industry, tremendous amount of data there that comes out of those logs every time you bring a car in for maintenance on that. And we have to look at that now. It's no longer the oil gauge. The times have changed. And these applications are fueling an avalanche of archival data. So you've heard some really great commentary and understanding of unstructured data growth. And we want to distinguish real carefully between structured and unstructured. In 2020, we see about 20% of all data fits in a database, and about 80% doesn't. It's just let's go find it beginning to end and try to find things. So there is this challenge. We'd love to see more data become structured somehow or at least semi structured so that we can find things. And the accessibility here trend is beginning to happen as more and more companies, and this is a software effort all the way. More and more companies are putting metadata catalogs, tags, indices and trying to create those on ingest when the data is actually brought into the system to have elastic archival data search capability, not only to expand your search capability very quickly rather than going back into the archives, and applying data post creation time. So this is a key trend as we begin to build our model for the 100-year archive going forward. Object storage is a key piece of this whole thing. And we can look here on this chart and see that object storage is very important to the archived game because object storage take files, they take videos, they take audio components. They put it together. And with that, they tie into that package or that object, the indexing capability, the tags, the metadata, some classification data, and that whole object now becomes a self-contained component to go find things and do some analysis on that data in an archive. It's not real good for online transaction data because you have to open up the object and analyze all that data on the front end. So that world is still going to be in the block and file world, mainly in the block world. But for archival data, 1 message for you to take away today is that the company that deals best with object storage and unstructured data is going to play a big game here. And when you put unstructured data and make it stores, allow it to be stored as objects, you create what's called semi structured data, which we see up there at the top. Okay. So there's a shift, I hope it continues. I really do. But the shift is continuing because we see that 80-20 split here that if nothing happens, it will still be the same as we saw earlier by 2025. But there's a gradual shift underway because so many people. And you saw that in the Quantum conversations this morning, we realize the need to add some structure with object capability to unstructured archival data. So we're looking at 20% now, and maybe we can drive that up 30%, 32%. This is a chart that IDC has done to project the amount of data that if we stay on this trend to help make data more structured and more searchable, we can unlock the archives and find data and unlock the unknown potential inside of that archive. So the trend is in this direction. It's going to take companies that have a great vision about how to do this and implement this in an archive strategy to make this happen. But once we put some structure with object storage around the archives and make it semi structured, we can accelerate maybe over 30% of the data by that time and will be 30-70 percent split. We'll have some structure where we can go find things within that. We need this to happen. Otherwise, it's going to take forever to unlock these archives as they continue to grow at 30% a year. So I want to put all this together into a strategy, and it's really the anatomy of this hundred year archive architecture. There's a lot of building blocks in this strategy. And I think it's key, the first 1 up there, and it's in red because it's the key component to make this entire archival strategy work is software. We need software, and it will require the kind of software that you've heard about this morning that's going to bring intelligence to do the things listed below this. We're going to need an intelligent scale out software engine to geospread unstructured and object data, which is unstructured and semi-structured data to manage these mixed workloads and let people get at that. We're going to need analytics and insights to understand better how to manage this. Because when you get into the petabyte, petascale-s and then exascale storage systems, the rules change. And things get a lot tougher when you get out there. The scale of this manager, you're in billions and billions of files. So the numbers make analytics and insights and predictive analytics, very, very important. Data classification on ingest, creating tags and metadata and information about the data coming into the system is a key piece of this architecture going forward. I love the strategy. Because these pieces allow you to bring intelligence with the data once it's stored into this architecture. Elastic search capabilities to protect data, like to find data quickly, take advantage of the objects and to bring this back to your business platform in a shorter period of time in the days and weeks it might take otherwise, if this doesn't happen though. We need an optimal archived technology. Bruno, I think made a great case for the reason that high capacity tape subsystems are the obvious answer to archive today. They have the lowest TCO really by a factor of 7, and its total cost of ownership roughly 1/7th of the disk system for the same amount of data today. I've done several studies on this and comes out in that range almost every time for archival data. You've got scalability with disc, you scale by adding disk drives with taping scale by adding media, you don't have to buy drives. So you can scale capacity much easier in the tape world here. Reliability for tape based on data rate has exceeded that of disc. Who knew. Who on this call knew that tape reliability was higher than disk. And that happened. It's got a 3 order of magnitude big higher reliability than disk. Media likes 30 years or more for modern tape and the RAIL architecture and absolute Quantum for pioneering this wonderful concept for rails here it's an additional availability protection for the archives. When you could use tape, you get another data protection advantage called the tape-air gap. And the air gap has basically all the cartridges in the library are electronically disconnected, so they cannot be hacked. So this is hacker-proof storage. You don't get to have data when it's off-line. There's no electrical connection. In the air gapped concept, which today only applies to tape in the data center, maybe there will be some other emerging technology in the future that will come in and also use that, but air gaps are key as part of the cybercrime prevention issue. And finally -- or not finally, but the next or the last step is to geospread this architecture. And geospreading is a really cool concept. I think you're going to like to see for archives because the software spreads data across multiple geographic locations using these RAIL architectures that you saw earlier, we had library spread arm. It breaks the data in what's called ratio codes or chards, little parts, portion of the data filing spreads ever across the geography. So if you have a geographic failure, an electrical failure an earthquake or flood, the data spread in other locations, and a minimum of 3 is usually the recommended way to provide maximum availability in case any 1 of those goes away. I mean you theoretically now don't need the backup function anymore to do this. So geospreading is just beginning in its infancy to be a key piece of the archive architecture of the future. But that is a wonderful piece and a highly beneficial piece of the kind of archive that we want to see in the last 100 years or longer. And then finally, nondisruptive data migration between technologies. You can't do this manually anymore. This task falls on the shoulders of software to determine what data needs to be where at the right time. And it will involve software that we haven't seen yet appear, and this will involve artificial intelligence to make predictive decisions based on past reference patterns as to how to make this data more accessible. So the brains behind this architecture is software. It requires a robust hardware infrastructure to be at the location where the data ultimately has to reside. That's the foundation of the architecture and you want to spread it across many locations and apply artificial intelligence as fast as it can come in the future to make this happen. So we will get into the final couple of comments here, and that's -- this Hyperscale comment. You've heard Hyperscale mentioned often this morning, but the shift in the storage industry is, we have fewer data centers today, but the ones we do have are much, much bigger. And this is called the Hyperscale movement right now. Hyperscale Data Centers are enormous data centers. Most of them are 400,000 square feet of race floor or bigger. The biggest one that we know is as big as 18.3 soccer fields, 1.1 million square feet of race floor space, unbelievable. The energy challenges here are staggered, most of that energy consumption comes from servers and hard drives. The major cloud providers are all Hyperscale Data Centers. Those providers listed there have over half the worldwide cloud infrastructure market to themselves. If we look at the data centers around the world, we had a 416 terawatts of energy consumed by these data centers last year, 40% more than the entire United Kingdom. And at least 45 data center -- each of these companies got 45 data centers themselves all over the world at a minimum, some have more than that. So on our chart, upper right-hand chart, you'll notice, by the end of this year, the projections are as many as 570 Hyperscale Data Centers. So this has become a critical piece of the storage spectrum going forward because people are putting their data in the cloud. They're not storing it in small data centers or in on-premise as often as they were before. Enterprises are still here, but the smaller data centers are saying, why bother, I'm going to go somewhere else. So if we look at what the Hyperscale projections are by the end of this year compared to 2017, in 2017, there were 386 known Hyperscale Data Centers, by the end of this year, about 570 million. By the end of this year, 47% of all data center servers expected to be in Hyperscale locations. We have some of these -- one Hyperscaler has got over 150,000 servers alone. 68% of all data center processing power consumed by Hyperscale. Over half of all data center traffic and 57% of all the data stored in data centers, 4.2 zettabytes, unbelievable what we've got going on here. So a little example below this, just for you to see the impact of using the architecture that Bruno described here. If all the Hyperscale Data Center archive data, which would be 60% of our -- or 57% of our number, 4.2 zettabytes. If that was all stored on disk, we'd have 28 million, 15 terabyte disk drives and 1.7 billion watts of electricity. If you put the typical split of 40% on disk and 60% of that data, which we said earlier, 60% of the data is archived on tape. You would see that the total energy savings would be with that kind of split by moving the inactive data off to tape about 945 million watts if tape is used for the archives. And this is a key trend for hyperscalers. They need the 100-year archive strategy yesterday. They need tape to relieve the tremendous pressure on massive disc farms that are consuming so much electricity that managing the energy component of Hyperscale is absolutely paralyzing. So for Hyperscale Data Centers going forward, physically scaling beyond the exabyte level will be nearly impossible without Geo-spreading architecture for archive and the use of tape. So the takeaways from this conversation today is, one, that data is the most valuable asset and is critical for modern business survival. We have passed that milestone long ago. We're looking at 60% to 80% of the data by 2025 to be archived. 4.2 to 6 zettabytes of data. It's the fastest-growing data class. Archived data preservation requirements now regularly exceed 100 years. You've heard that throughout this presentation today. Making archives accessible, metadata at ingest, tagging, indexing, classification. This is something CIOs have to have to make any use out of these archives. Converting unstructured data to object data with more metadata and tags to make it at least semi-structured is a huge plus in this example. Hyperscalers, pushing the limits of every architecture that we know from compute to storage to the physical infrastructure to energy, unbelievable. This is a huge pressure release valve on the massive constraints for hyperscalers of keeping large amounts of inactive data spending on disc 7 by 24 by 365. The responsibility here falls on advanced software, and the companies with that vision will be the ones that succeed mightily in this space as we see how big it is, but advanced software, data classification, geo-spread, erasure coding, the company that can master and integrate those components gets to win the prize in the massive storage challenge that lies ahead. So this is a big challenge. We see in a way to get -- make this happen, and it's exciting for me, since I live in the middle of this space to see all the great things that are beginning to come together in a very crystal clear vision. Thank you.

Thank you very much, Fred. It's always learn a lot listening to you. It's always interesting to get your perspective. We'll be publishing a couple of white papers with additional information based on what Fred discussed. And Fred, I actually think you and I will be doing another webinar in September to talk more about this market. So we've heard a lot about Quantum technology. Now to talk about how we're going to bring that to market and how we're going to transform our go-to-market over the next few years. I'd like to introduce Liz King, our Chief Revenue Officer. Liz?

Thank you, Eric. And good day, everybody. It's a pleasure to be with you today. As you've seen from everyone who've spoke -- who has spoken now, we are outlining an ambitious plan to transform our company with new business models and a broader portfolio that gives us the ability to pursue and capture a much larger total addressable market. And with this transformation, must come a transformation of our go-to-market strategy, the way Quantum will go to market to deliver this ambitious plan. Go-to-market is how we reach, sell to and serve our customers. And most importantly, how we deliver the most competitive and compelling customer experience. So before I take you through how we're going to transform go-to-market, let me just take a few moments to tell you why I actually joined Quantum. So I joined Quantum a little over a year ago. After working for many technology companies, and I joined the company for several reasons. First, our vision, you've heard a lot about it today. It's exciting, it's relevant and it's positioned for dazzling growth. Second, sales transformation is actually what I do. Every role I've had throughout my career has been one of sales and go-to-market transformation. Whether it's building, rebuilding or evolving or a combination of all 3, our customer-facing organizations to deliver profitable revenue growth expand it into new markets and do all of that with operational excellence. We also have fantastic customers and partners. That was amazingly compelling to me. You heard from 2 of them today. They have an incredibly large and massive challenge ahead of them, managing all of this unstructured data coming at them and growing exponentially. And so it's a perfect place for Quantum to be, to address and rise up to those challenges. And finally, we at Quantum have a customer-driven culture. We are maniacal about the customer experience. And all of these areas here, which propelled me to Quantum, it serve as a fantastic foundation upon which to drive our go-to-market transformation. So let's talk about that. So it's built on 3 pillars. The first is what I call solutions-based selling. This is moving to a solution-based selling model, armed with a broader and richer solutions portfolio that we mapped out for you today. The second is market and geographical expansion, propelled by that enhanced portfolio. And then finally, operating a best-in-class sales machine, I call it a sales engine to deliver on this ambitious growth plan. So let's first talk about the first pillar. Solutions-based selling. When I joined Quantum, it was very clear that we were great at selling our products. That was wonderful. It just limited us as we focused mainly on the transaction and not necessarily the entire outcome the customer wanted to achieve a bigger, broader picture, the longer-term strategic outcome. So we have evolved. We've evolved to an outcomes-based selling model. Changing the conversations with our customers to anticipate their business challenges and truly help their -- them achieve their mission. This is a critical behavioral change that enables us as a company to be a better partner, a better strategic trusted partner to them and deliver a sustaining value to them over time. Many of the new offerings that we've introduced actually came from these kinds of conversations and feedback that we've received from these customers with their real-time pulse on the market, we've been able to pivot to better respond to their needs. So much of what we've talked about today is we're through those conversations and continue to this day. And hence, we now have a vision on selling a broader and modernized portfolio with subscription and services-led offerings, relevant to the cloud, aligned to the cloud. This will deliver long-term sustainable -- it will -- long-term sustainable value and will deliver a great customer outcome. And so what does this mean in terms of results? Well, first of all, it means a deeper and more profound relationship with our customers as a trusted partner. We're deeper, and through that depth, we now are able to deliver larger total contract values, not one little transaction, a much larger, longer-term solution-based contract value with them. Larger average selling prices through just the conversations we've had since I've been here -- are -- the size of our contracts are growing. And it's because we're selling more. We're stitching together our portfolio in different ways. We're wrapping services around it. We're listening, and that creates an expanded footprint into these accounts. And then with subscription and as a service offerings that we're developing, we will no longer just have episodic onetime transactions. We'll have also, through all of this, increased margins through a much richer revenue mix and ultimately and basically greater economic outcome. Let me be clear, we are selling solutions today. We have blended our current portfolio, as I've said, and we're capturing more share of wallet and our customers, both in our traditional markets as well as new markets. It's a great start. It's just the beginning of the journey. So let's talk about now the second pillar, which is expanding into new markets and geographies. So we see expansion. If you look at the slide, it's a -- think of it as a 3-dimensional continuum. So now we're armed with the richer-based -- value-based portfolio and a solutions selling mindset that just evolves and grows as we expand our solution set. We will expand then into new verticals, broaden our reach into the geographies we serve today. We've been very strong, as you can see in the middle, in our traditional markets of the media entertainment, especially in the Americas as well as backup for the enterprise across all of our geographies, providing best-in-class storage infrastructure for managing video and unstructured data. So -- and we've already been able to, beyond that traditional core to capture new markets. This expansion, as our solution sets expand, will become much more profound. So let me give you some examples as we go down the left side of the dimensional continuum. First, with government, we serve many government and public sector customers today. We see this massively growing because data is growing, and they have the increasing need to manage and maintain this data for very long periods of time. We also see the edge, we call it the edge, as Ed mentioned earlier, for data collection, analysis and portability, video surveillance across the enterprise as well as autonomous solutions for automotive, autonomous automotive needs, self-driving cars. But even beyond that, think any vehicle, any type of mode that collects images and video. So there's much more capacity for that. Also our cloud providers, we're serving the top hyperscalers today. It's an awesome, awesome -- and an honor to do so. There's also more just beyond the top of fast internet companies, net services providers, any services company for which data collection and secure archive are critical requirements for their own business models. And then finally, life sciences, health care and research, where effective data management is critical to accelerating both data collection analysis and most importantly, discovery. We're selling into all of these areas today. It's just now with this enhanced portfolio, it will be much more profound. And that takes me to geographical expansion. Today, we sell into 120 countries. Across these 3 regions, the Americas, Europe, Middle East and Africa and Asia Pacific. As we open up into new markets, we then open up into and expand into new territories. To extend our reach within our customer base, of course expanding that footprint, but also beyond that customer base. We are strategically investing today to accelerate this expansion, and that takes us to our third pillar which is sales engine transformation. So what do I mean by sales engine? Well, think of it as humans and machine. It's really a combination of our Quantum sales force, our channel sales organization, which is very large and the machine, the operating model that basically enables the selling motion. This sales engine of humans and machines has to adapt, too, to our transformational growth, to adapt to the new business models we're introducing. So that we can actually expand and grow across those 2 dimensions I talked to you about. So let's first talk about the Quantum sales force. So when I started at Quantum, my focus was certainly on stabilizing the business, understanding and assessing our skill sets, which are very profound and very strong. Optimizing everywhere we could, make rational sense out of our spend, and then deliver our numbers, align operationally and organizationally to our vision and then bring in top talent and address gaps where we have so that we can build for the future. Think of it like the old expression, not skating to where the puck is going, being where the puck is going. So since then, 70% of my leadership team is new to Quantum. They bring new and deep industry expertise, solutions, software-led, services-led expertise, they have experience and diversity of thought to enable us to further expand into new markets and geographies to help propel our existing sales organization to evolve. Also, 15% of my entire organization is also new to Quantum, bringing that same level of expertise, vertical knowledge and also geographical expansion, as we augment the rich foundation of the company that we have today. Now we've made these enhancements in a very optimized way. And as we grow, we will continue to build, enable and augment this capability. Now this is one side of the coin. The other is channel, our channels transformation. We have a very partner central -- centric model at Quantum today. The majority of our traditional sales has -- today, goes through the channel. For this reason, we have to evolve our channel just like we're evolving ourselves. And so we brought in a new channel leader, earlier this year, with deep software and solution services-led experience to completely revamp our channel program. Again, he is where the puck is. Where the puck is going, he's already there. So we're going to be revamping our program. It includes enabling our partners, training them to sell our enhanced portfolio, but it also includes recruitment. Basically recruiting new partners who are already there, who are well equipped to sell this broadened, new and modernized portfolio and reach across new markets, extend our ability to penetrate that really huge TAM we're going after worldwide. This revamp program is already under construction, and it will roll out very soon, so stay tuned. And finally, go-to-market transformation requires operational excellence. You have to have a very well-oiled machine to run your business. You have to be able to manage what you can measure. So metrics are very key. So we've set up a new sales operations structure, focused on sales productivity, sales automation and speed. The structure has been designed to ensure that our sellers, whether it's our sales force at Quantum or a channeled army is enabled and trained to sell our portfolio and to adapt to these new models as we move to sell for subscription, services led, Quantum-as-a-service. All of these new revenue models, it is where we're going, and we are already designing and adapting to that. And also, this is something I've always experienced in leading sales teams. You want to spend as much time as possible every minute of the day with your customers. Speaking with your partners, collectively getting feedback, having that strategic and trusted relationship. You don't want to be burdened by old tools or just inefficiencies operationally. And so we want our teams to spend more time on those engagements and build that richer revenue pipeline. So this structure, what will it do and it's currently underway, will deliver greater revenue and margin contribution per seller and per channel partner. We'll measure this very tightly. The automation that we're going to be introducing as well, will bring a lower cost per customer transaction as well as efficient online customer experience and we're maniacally focused on the customer. So we're going to move to this online experience. It will be much more enhanced than what we have today for customers to not only self-serve, but receive immediate response, have that 2-way dialogue very quickly and again, drive speed into every transaction we drive. So it's underway. It's under construction. We're really excited about how this technology will drive greater efficiencies and optimization to improve our overall economic fundamentals. So in summary, we have 3 pillars of our go-to-market transformation that we're driving right now, solutions-led selling; market and geographic expansion; and the transformation of our sales engine. All of this leads to the capture of more than $75 billion in total addressable market, that's available to us, what Jamie shared with you in his opening. This will be available to us, and with it will come increased margins and recurring profitable revenue with our modernized portfolio. So I look forward to sharing updates with you as we drive our go-to-market transformation in the months and the quarters to come. So thank you, and Eric, back to you.

Thank you very much, Liz, and just to talk a bit more about what this means for our finances. I'd like to introduce Mike Dodson, our Chief Financial Officer, to walk through the financial model. Mike?

Okay. Thank you, Eric, and welcome, everyone. We really appreciate you taking the time today to give us the opportunity to provide our strategy, our vision. I know many of you, Jamie and I have spent time with over the last couple of years, typically in 30-minute segments. And to give us an opportunity to go from start to finish and really bring everyone up to speed in one session. Hopefully, you're finding the great value today. And before I give more color on our long-term model, that Jamie has already introduced you to some of the key metrics. I'd like to first be able to provide what have we accomplished over the last couple of years. And then also talk about our capital strategy -- our capital structure strategy. So starting with our financial progress to date. In December of '18, we refinanced our debt. We were in the midst of a multiyear restatement, addressing SEC issues, liquidity issues. So we are very fortunate to be able to get a favorable refinancing at that time. And then with the lenders that we did that deal with, most recently, addressing the issues with COVID-19. We amended that agreement. So we have a very strong partnership with our lenders and we'll go into more details. But we believe we've got a more flexible capital structure today. We've also improved our operating model. We've taken $70 million out of our operating annual run rates, and that's out of both OpEx and cost of sales. And a lot of that is just locking and tackling, turning over every rock that we can. But we feel like we've made a lot of improvements there, but we still believe there's more that we can do. We've resolved SEC matters, related litigation, which then allowed us to relist on NASDAQ. And then the sum of all of this, when you can look at these 2 columns when we compare fiscal year '17 to fiscal year '20, we have doubled our EBITDA, it was 20% less revenue. Now as we go down and we look at the components of how we did that. The lower revenue is, as Jamie has described and Bruno has described, the backup tape is a golden glide. So we've got that headwind from a revenue standpoint. Also in this revenue decline is royalties, which were down $20 million between these 2 periods. And that also, not only is that a revenue headwind when you talk about royalties, but that's a gross margin headwind, as that's a 100% margin. But even with that headwind, you can see we improved our gross margin. And the improvement in the gross margin, really, as Jamie had mentioned earlier, and we're focused on more value products out of the reselling products, which are lower margins, less value. And of course, we reduced our costs as well. So a little better on margins, but we had headwinds that we had to come over. But there already, the biggest leverage that we've gotten when we look at what we've done over the last few years is on the operating expenses, where we've improved as a percentage of revenue, 7%. So that's the real leverage that we've enjoyed. So you can see that comes down to the operating income. We had 2% in fiscal-year '17, and we did 5x better, 5x to get to 10%. So it's double digit. And that was, of course, fiscal-year '20, that ended in March. So we had a little bit of the pandemic impact to that, but not a lot. And then the adjusted EBITDA, you can see more than doubling when you had that big of an improvement in operating income. And then the EPS leverage that you have, we got $0.34. And all of these results we're looking at are on an adjusted basis. So we take out any special onetime activity as it relates to litigation or restructuring. They pulled out of these numbers. And there's a reconciliation later than that, if you're interested. So that's what we've been up to, the last few years. Now I'd like to just talk about our capital structure a little bit. The first, just, what do we have today? We've got a $45 million revolving credit line. That's with P&C, they've been a very good partner for many years. On a term debt basis, we've got $185 million today. The lead underneath that is Pinnacle and then with Blue Torch. And then we've got a $10 million PPP loan as well. So that's $195 million. When you look at that, excluding the PPP loan, it's 5.6x our trailing 12 months. So our trailing 12-month EBITDA, $33 million, includes this Q1, which is at the trough of the -- of dealing with the pandemic business pressure. So we're at a higher multiple than we would like to be. Really, we'll target 2x to 3x trailing 12 months. And we'll look at our strategy here in the second. But obviously, today, we're not in a place that we feel is optimal by any stretch of the imagination. The term debt today is at 12% which is 10 points over LIBOR. The LIBOR has a floor, 2%. So we're at 12. And it's a payment of $5.6 million a quarter. So almost $2 billion a month. When you look at our covenants, working through the amendments with our lenders. We've got a levered through March 31, 2021. So again, a very good partnership there. And we've deferred our amortization payments through the end of the calendar year in December. So when we look at the prepayment terms, we have the make whole, which by next summer, the end of June is in place. And it required a 7% premium call, call premium on top of the interest that we would pay through June. So it's really -- it's a very expensive proposition to look at prepayment until we get past that period. But exclusive of that, we can do an equity call back, if the stock prices did make sense, we obviously do not want -- we would not want to dilute our shareholders, but we do have the ability with a 5% call premium to do it an equity offering and take down after 15%. So we do have that. Our stock price was much higher. If it made sense, we wanted to be opportunistic. We do have that opportunity. And then the call premiums, following the make whole for 6 months until the end of 2021 at 7% in the fourth year, our 5-year deal is 4% and then in the last year, through the -- for the calendar year 2023, there is no call premium. It's a pretty thing. So that is what we have in place today, just to give everyone the terms. When we looked at what is our strategy. The first thing we'll do is we'll apply, and we'd expect to qualify for the forgiveness of the PPP loan, and that's 100% of the $10 million that we have. Then when we looked at our make whole terms, really, by the end of June, so next summer, we want to work leading up to that date to figure out what's the -- what are our options. We have -- we believe, we have a number of equity capital market activities that we've looked at. But really, the -- we love the partners we have today that really help us in a very difficult time of the interest rates we pay today, we think we'll be in a position to do much better. So late next summer, we would expect to be addressing our current structure as it relates to our term debt. Because really, as you can see what our target is, 2x to 3x, and we want to be at more normalized market interest rates and confidence. So those are our thoughts today as far as the capital strategy. Now the 3- to 5-year financial model. And I'd like to -- Jamie has gone over a little bit of this, but I'd like to provide more color. First, we know underlying this model is the assumptions of the business model transformation. And these are not insignificant to go to a software-defined subscription model, to go to a storage-as-a-service, which we're calling Quantum-as-a-service, to automate our service delivery, all of these are significant business transformations that we believe will take years. And on top of that, we're looking at TAM expansion. We've got new verticals we can go onto, as Liz had outlined geographies, we want to be more aggressive in EMEA and in APAC. And we've heard a lot about the 100-year archive requirement. So we think there's a lot of opportunity on the TAM expansion. And of course, we're in huge markets, and those markets are growing. So we think we're in the right spot. So what does that mean to our business model? First, you could see are currently the split of our revenue mix. Roughly speaking, we're 60% in product or 40% recurring revenue. The recurring revenue today is our service revenue and then no royalty. What we're looking at through this transformation is to have 30% products and 70% recur. So that split, when you look at what is going to still stay as product, that's principally the Hyperscale business there. We'll probably always be in a position that we're selling them product. We're selling them equipment. But everything else is going to go out as software or it's going to go out as a service. So 70% of that is going to be very high margin, 30% will be much lower margin. What that blend is, is 60% total for the company. So at 60%, that's a significant improvement over our current low 40s that we see today. And then when we look at the operating expenses, we really -- we want to continue to get leverage there. We won't get as much as we did in the last few years, but we're very disciplined in this area. We'll continue to spend the R&D as a percent. So we'll keep our investment very steady. So as we grow our top line, we'll grow our R&D, but we want to continue to optimize sales and marketing and G&A. And we think we've got opportunities in those areas to continue to get more leverage as we get scale there. So we get a little bit of benefit in our new model, but it's not significant. But we're not -- to say it another way, we're not going to let that go up, right? We're not going to lose leverage on that line. It will be very disciplined. What does that mean from an operating income line? That's 30%. So if you're making 60% gross margin, you got 30% OpEx, that's 30%. So when you look back to where we are today at 10%, 10% feels good when you look where we were at 2% and getting to double digits, but we're 3x. So we're at 30% operating income levels, that equates directly to 3x EBITDA, right? So if we're bouncing around $45 million, $50 million EBITDA currently, that's $150 million of EBITDA. So it just shows you the leverage we have in the model. And the power that a recurring revenue model has. And what does that equate to EPS, which you get even more leverage with at $0.34, we're going $0.06. So that's plus or minus $2 this year. So when we look out 3 to 5 years, maybe that seems like a wide range. But with all of this transformation, we think it's realistic to give it that kind of a range. But we believe it's a totally new business model. This is not a product based, grinded out hardware, compete with a lot of big guys and make margins in the high 30s, low 40s. We're looking at a blended rate of 60, and we should be able to do better as we move forward beyond this period as well. But it's something that has got great leverage in it. We've got great products. We've got technology, as you've heard today, we're in great markets that are growing. So we're really very excited that we have everything to be successful. And this model will look a lot different, as we go through this transformation. And so that is all that I have today, and I'd like to turn it back to Jamie for closing comments.

Thanks, Mike. I think, we've laid out a pretty bold plans for the future. But I think we've underpinned the plan with core execution. I think we feel comfortable that we can deliver on the products and technologies that are needed to carry out this transformation. We've done testing of the market. I think our customers want us to carry out this transformation. And I think it's the best thing, not just for our customers, but I think it's the best thing for us to not just stabilize the business, but begin to move into a higher growth and a more powerful earnings model. You're going to be seeing a series of releases, our next version of StorNext Software-Defined, you'll be seeing releases of StorNext available on the cloud. We'll be putting out actually a new cloud system in the next several months that has very different characteristics than StorNext, and we'll be moving into not just new products, but these newer products will be only available in the new business model. We're not going to offer the customer multiple choices. Our newer products will only be available on subscription. Some of our platforms are only available as storage out of the service. So we'll be not just introducing new technology, but introducing whole new business models as we go forward. And should be pretty exciting. We're going to learn a lot, but these are just transformations that, from my point of view, are required. There's simply requisite to be relevant in today's tech marketplace. So with that, I'm sure we stirred up a lot of questions, Rob or Rob and/or Eric, why don't you lead us through and moderate the Q&A session.

Yes. Thanks, Jamie. [Operator Instructions] We're going to start off with a question from Craig Ellis of B. Riley. Jamie, as it relates to the $76 billion total addressable market and the weighted average CAGR of 18% that you overviewed on Slide 10, can you discuss the multiyear TAM segment penetration rates? And whether the 18% CAGR is a good proxy for Quantum revenue growth along the notable gives and takes?

Yes. I mean, the TAM is so big in comparison to today's revenue. I think to be realistic, we only need to pull off a few points out of that TAM to get a much more powerful result. So that's a big part of our calculus is the markets we serve are so large, the need for 100-year archive is so large, the size of hyperscalers is so large. Again, we don't need to get 80% market share to be successful. A few points drive a big outcome for us. As it relates to growth, we're going to have to be thoughtful because we're making a transition, meaning today, when we close the sale, we rev rec 100% of it minus the support contract right upfront. In our newer models, we're going to be recognizing revenue over the period of the contract. So 112 per month, 124, 136, depending on the term of the contract. So it's possible that our revenue is flat or down, yet we have a huge increase in total contract value, a large increase in deferred revenue, as we begin to prepare for more of a recurring revenue model. Now we're going to have to manage that transition delicately as we go from hardware to as-a-service. And that transition will take some time. So it's not to disrupt our business model. And we don't know exactly how that's going to play out right now. We know recurring revenue is better for us. We know software margins are better for us. But this transition exactly how it's going to play out. I think post the transition, I do think this is a double-digit growth company, but we got to get through the transition. And it's likely that won't be completely smooth. I don't think anyone's been able to make these transitions in completely a smooth way. But again, my point of view is this is a transition that just must occur.

Okay. And along those lines, you did touch on this, how linearly can business track from a current level to the target levels? Acknowledging that the long-term strategic update and recent customer engagements, deal size, funnel applications, like how does this all play into sort of 2022, the covenant entrainment and the long-term financial model?

I think, I don't have the covenant concerns about this. I just don't think that enters the calculus. I think we have to play this out. And we're going to do it in a structured way. We're going to do it in a methodical way that we carry out the transition probably over 18 to 24 months. We still had a large install base that's on the more traditional model, and that they will be with us for many years. We have customers who bought from us on traditional models that will still be with us for many years. So it is a well-orchestrated transition. And the key -- the A plus that we're looking for is to be able to make this transition and increase revenue, which is really hard to do when you go to more of a deferred revenue model than an upfront rev rec model. So that's the ace -- an A plus would be we make the transition and revenues stay flat and even go up. As you load up deferred revenue with more long-term contracts. So we're going to have to see how it plays out.

Great. And when you think about the software and storage-as-a-service transition, are the early adopters and fast followers more prominent in particular verticals?

I don't really think so. I think everyone -- I think the entire IT consumer community whether it's moviemakers or genome sequencers or university researchers or traditional retailers and hospital administrators, they all have a similar set of careabouts. The first is, I want to pay for an outcome. I don't want to pay a huge amount of money upfront and hope that your product does what you say. I think they like the, I pay for an outcome, I pay for a service, I pay for something and I get that service. I think IT consumers are moving more to this, pay as you go model. I also think they're moving to more elastic models. And sometimes our needs go like this, when we buy a lot of hardware, we have a set need, and they're moving more to these models of my needs go up and down, and I want your service to be able to go up and down. I think IT consumers are moving more to a software-based value model. It's the software that gives the features, the capability, that's a software that differentiates, they want to put more of their time, money and effort into that software. So -- and I think that's across the industry. And I think they're looking for more solutions-oriented approach and what Liz meant in that was customers saying, if you sell me a box and you drop a box off at my loading dock, it doesn't mean I'm going to -- if you're NASA, it doesn't mean you're going to get to Mars. It doesn't mean you're going to build a moon base because you just dropped off a box at my loading dock. It doesn't mean we've made a bank more successful. It doesn't mean we've made a retailer more successful because we just ship them a box. What we're moving to is, we're sitting with our customers and saying, what is it that you need to get to Mars, right? You need high fidelity photographs of the surface of Mars. Okay, how can we help you do that? If that success of the mission is, we're going to take incredibly detailed photographs of the surface, of rocks, of the atmosphere. Photographs from above the atmosphere, below the atmosphere, if that's a definition of success, that's what we want to deliver to you, not just a box. And I think every -- all IT consumers are thinking along those lines, and that's both a way of engagement with the company and a way of paying a company. I want to pay you as I'm successful versus paying you a lot upfront. And then it's just on me to get to Mars by myself or make my banks more successful where we're partnering with them in a deeper way. And that allows us to get more of the TAM. It allows us to build a stickier relationship because when we achieve that, what it means technically is they're stitching their business to our products, they're calling APIs. They're calling functions, we're customizing our software to their mission. So when they do that, our product is woven into their company's fabric for 5 to 10 years instead of, "hey, we used your product for 3 years then it's depreciated and we chuck it." It's a very different model. Our software has woven in for a decade or more. And it's proven really successful.

Great. Next question comes from Eric Martinuzzi of Lake Street. Fred's presentation overviewing the archive growth of 25% to 30% is compelling. How should investors reconcile that versus the company's revenue trajectory over the past few years?

Yes. The last few years, it's been somewhat miraculous, right? I mean we were days -- we were weeks away from closing our doors. And in less than 8 quarters, we're in a radically different place. So I think if we stay on the trajectory we're on, it's going to be pretty explosive. Just given how much we've been able to innovate, we've been able to move the economics of the company, we've been able to attract talent, we've been able to hold on to our customers. The incredible success moving in the #1 position in hyperscale archive, and we are #1 in the position. We are #1 in high-end post production in television, movie and sports. We are hands down #1 in that space, and we're pulling further away from the competition. So prior to COVID, it's pretty amazing. We became profitable again. We are growing again. We are GAAP profitable, cash flow positive. And no one prepares for a global pandemic. That took the wind out of our sales, a 30% decline in revenue is pretty tough. Big exposure to the media and entertainment space. But we announced, hey, we did 73. We called 83 at the midpoint. I feel very comfortable on that, and it's probably going to come up again as we have a double effect happening to us now. We have growth because we're coming out of the pandemic, and there's more normalcy or more adaptation to the pandemic. Less surprised, more adaptation. People are just learning to move forward under these conditions, and we're having growth happening as well. Natural growth from our inorganic activity from the new products we're releasing, and we are sitting on the cusp of the largest series of releases that I think we've ever done in this company's history. Brand-new StorNext products, we're going to be launching another file system, something that is pretty exciting along with StorNext. We're going to be launching our RAIL technologies new versions of Active Scale. We have a new all NVMe version of the R-Series. It's a pretty exciting time for us. And what we are seeing in the demand building, I think we are laying down the train tracks for sustained growth to be able to power us through the transition, which will probably put pressure on revenue not that we book any less. It just we'll recognize less because instead of it all being upfront, it's going to be ratable, again, which I think is just a better business model for us.

Great. And you just mentioned RAILS. Craig Ellis brought it up. It seems like an important evolution for hyperscalers and large enterprises. He wanted to know when it became available and what's the initial hyperscale and on-premise enterprise reaction? Can you discuss the systems, robotics and media pricing?

Yes. I mean we're not announcing any pricing, but the pricing model -- the pricing model for RAIL, which is basically an object storage, a very sophisticated object storage technology that parses data across tape and disk to centers all over the world. So it's a geographic spreading of data, compression of the data and then putting it on formats of disk and tape to provide different speeds of retrieval and placing it on that tape where it could last for over 100 years. So being able to actually change the underlying hardware over the course of 100 years, but the data is spread around it, is designed for that. It's a pretty incredible architecture. We are finding certain companies -- and I think many companies are building their strategy based on what is called a hybrid and multi-cloud approach. I think most companies know, they're not moving 100% to the cloud. They'll have data on-premise, they'll have data in the cloud, and they've got to decide when it goes where. And I think the RAIL architecture is going to sit and many of the world's largest companies are going to have 100-year archive. Think of it this way, if you're going to keep data, let's say, feature films, which is an immediate example. If you're going to archive them for 100 years, are you really going to pay a monthly fee to a cloud provider for 100 years. It just seems -- you'd say, "hey, I'm going to be paying for 100 years. I want to go set up my own infrastructure for that." And then still may put some of the data on the cloud for others to gain access to, to share it in certain ways. I'm still going to use the cloud, but my last step, my last defense of protection it's likely going to be an on-premise environment that's completely isolated from ransomware. It's -- the network is locked in. They have total control of that. The other area where cloud gets a little impractical is if you're retrieving often. The exit fees are extraordinary. So if you were to put all your data in the cloud and one day decide that I'm going to move it from cloud vendor A to cloud vendor B, you would have to pay exit fees on every single file you put there. Those fees can be tens of millions of dollars, right? It's designed like a roach motel, where you check-in, you never check out. If you were to actually put a ton of stuff at one cloud vendor and try and move into another, I mean the bill shock is just extraordinary. And I think a lot of people just don't want to place themselves in the corner. And that's not to say that the cloud isn't going to keep exploding. It's just, I think, you're going to see organizations have both on-premise technologies and cloud technologies, and they're going to work together in a hybrid way. I think most CIOs look at the world that way. And I don't think our cloud customers are upset at us that we are still putting equipment on premise. We've been doing it for 40 years. We have every right to continue that business model. I don't think they view it as competitive. It's just, I think, like virtualization, when the world moved to virtualization, everyone thought computing would consolidate. Did it consolidate at all? No, we just did more computing. And when the world moved to the cloud and, "oh, everything is moving to the cloud." No. We're doing everything we ever did on-premise and even more on the cloud because the world is doing more computing. So we're simply as a society, the cloud virtualization, the technologies we're doing. They're not shifting stuff away from one person to another. All of us are just doing more. We're doing more on-premise. We're doing more virtualization, we're doing more on the cloud. Organizations are just doing more analytics, more computing, more data. It's not actually taking away from one person's pocket and going to another, we're just all doing more.

Great. And when you're looking at the software subscription transition and TAM expansion. How much is this is greenfield? And how much is vendor displacement? And what type of software tool would you be augmenting, displacing? And how do you think of the competitive landscape overall?

Look, I think there are multiple competitors chasing a similar vision to us. I think we are validating each other's strategies. I do think in data management in general or just storage in general. The emphasis is moving away from just storing the bit, protecting the bit. It's moving more to help me get organized. If you think about Global News network, they have clips of the news, interviews and news segments spread from their offices in New York, San Francisco, London, Hong Kong, Singapore. They've got data all over the place. And now post-COVID, it's sitting in every editor's house. Like this -- like getting that organized, where's all my stuff? How do I find my stuff? Hey, I need that video clip where best celebrity was riding that horse bet afternoon in that field. Like can you imagine how someone has to go find that. That's what the competitive landscape is moving to. It's not just store my bits, keep my bits there. It's what is it? What's the context of that? What is happening in that cloud? Where is that genome that was deformed and subject to this cancer? I've got to find that gene and that's where I think the competitive landscape is moving. Who can help companies understand, search their data and mine learnings out of that data. The companies that do that are going to be the most successful storage companies, and we're in an absolute all out sprint to get there before anyone else. And having worked in a lot of very large companies, Quantum's biggest advantage is we have velocity. We can move in this place, and we can move fast. I've worked at enormous companies. And when they finally get to their destination, it's amazing, but like when that battleship takes -- it's more like an aircraft carrier, right? It just slowly turns, slowly moves across the ocean. When it does finally arrive in theater, it's pretty awesome, but what I'm seeing is their arrival in our theater will be years late. We're just moving at a velocity that I just don't think our bigger peers can move at. And that's our advantage, and that's why we're moving so quick.

With that velocity in mind, question from Chad Bennett. How capable is the current channel to sell software-oriented products? And do you need to develop new channels for the incremental verticals and TAM that you're going after?

I think what we're talking about is not super new, right? It was Cisco over 10 years ago, who went to the hardware sellers and said, "Hey, folks, you're not going to make a lot of money just reselling commodity hardware. Right? Selling commodity servers, commodity storage, commodity, networking equipment, the channel had trouble making money in that area for some time. So most high-quality channel partners are really -- channel partners that are still alive and relevant, successful today, have already made the transition. They've moved to selling software-oriented solutions. They've moved to hyper-converged solutions. They've moved to selling cloud and hybrid cloud technology, and they've moved to more of a services orientation. So I think there just aren't that many successful box pushers out there anymore, and we're aligned to that more modern channel. I think some of our channel will be more relevant than others, but I think the biggest channel opportunity for us is new markets. The video surveillance market is a different channel. The archived channels in New Zealand, Australia, Singapore, I mean, there's markets -- I mean, let's be honest, I mean, we're a very North American company. And the other big effort we're doing is we're starting to put engineering offices. We're starting to put development offices into Asia because we want to be a global company. And we've got to break the fact that we're really good at selling in New York and Los Angeles. We've got to be just as good at selling in Hong Kong, in Macau, in Singapore, in Australia. We've got to learn to do that. And that's a big part of our channel strategy is we've got to be relevant to those markets and the channel partners in those markets. So I really think that's where our channel recruiting is going to be is in these new verticals. New verticals like health care, surveillance, autonomous. And we can't just be great at autonomous in Detroit. We got to be great in autonomous in China. In Europe, in Germany, in France, where cars are being made. And there's a lot of cars being made outside of the United States. We've got to be relevant there. So I think that's a big part of our strategy is being just as good overseas as we are at home and really becoming less of an American company and more of a global company. And that's a lot for us.

Yes. And Mike, just a question for you just on that channel theme. Do you have -- what percentage of sales come from channel partners today?

Yes. Today, the only thing that does not go through the channel is our hyperscale business and OEM.

We have a bunch of questions really about the long-term financial model that you laid out. So I'm going to try to pool them together. But Jamie, Mike, can you give any further context sort of narrowing down the general time frame here, where you're looking to get to the targets that you outlined today?

I don't think we're ready. We are putting those products that are more oriented to this vision into the market actually, many of them are in trial today. So we have customers using StorNext entirely in the cloud today. Our next version of StorNext is fully software-defined. We have customers that are trialing that today. We have new versions of DXi that backup their data directly to the cloud. We're putting that into the market and literally in several months or less. So these products are going into market. We've got to get a better understanding of adoption rate. We've got to get a better understanding of how our customers will plan this transition. And I think it would be foolhardy for us to be able to not in tidy 90-day increments lay out exactly how this is going to play out. Right now, we're making these transitions. We're committed to them. And as we make them, I think we'll get better underpinning metrics to give a strong model, but we're in business transition, and we're in the middle of a global pandemic. I think to give tidy 90-day goals. It's just -- I just don't think it's possible. And I think it could be misleading, and I just don't want to go there.

Yes. Appreciate that. And obviously, the transition between a traditional model and that is always hard to predict, but in your model, I mean, do you have sort of a breakeven that you're looking at in that? In other words, how are you thinking about the price here?

Well, the -- as we transform, yes, it's obviously -- it's accretive right out of the chute. I mean as we transform, we would be looking at gross margin dollars at the same level, if not more at lower revenue levels. So it's -- we're not -- there is much such easy on a breakeven analysis, right? I think it's all of it.

Well, we don't expect to go deep into the red to make the transition. We're not going to break the discipline. We're not going to stop spending wisely. We're not going to go into the negative. We're going to make the transition while maintaining our financial discipline. And that's the needle we have to thread, but I think we've been threading needle since the day we walked in here. And I think it's -- compared to the things we've had to go through to get here, I think this is an easier walk.

All right. I have some questions to incorporate some of your colleagues. This question is for Bruno. What secondary storage revenue mix across Active Scale, Quantum Scalar and DXi? How will this evolve over time overall and with hyperscaler and other customers? How will the service pacing process across these lines?

So I would say for my business unit, tape is still the biggest revenue contributor and then Active Scale and DXi are at even levels comparatively. And then in terms of implementation of the pay as a service or even starts. So I would say Active Scale will be leading in that area. And as Jamie mentioned, the -- our, let's say, on-prem cloud will be fully staffed and subscription-based when it rolls out later this year, right? So that will transition very quickly. DXi will probably trail there a little bit, but we're working on cloud implementation for DXi as well. So that's -- I think the -- needed there to make that transition happen on the DXi side.

And Bruno, another one for you here, and this is directed to Jamie as well, but given the obvious benefits TCO has, why are hyperscalers taking so long to commit to Quantum or other vendors for their archival tape investments?

Yes. I mean, Bruno, I can...

Yes, you can get solid, Jamie. I have some thoughts on this too. Go ahead.

Most hyperscalers have 2 bodies of work they have to do before they get started. The first is the larger hyperscalers want to write all of the storage software themselves and to write a ground-up storage stack or tape that is pull through, right? They cannot take a PR challenge because that lower -- if you remember the pyramid that Fred Moore showed the lowest tier of the archive here. Yes, it's the lowest tier, but it's the last line of defense. You can lose data in those other tiers because you always have one copy back down at that low tier. So they have to build this software and they have to test it to an incredible level because if you were to lose that data, there's often nowhere else to get it. So 2 to 3 years to build and test that is not an abnormal time frame. Then they have designed a set of practices and methods to keep that equipment running 24 hours a day, never ever going down for 1 second. And they have to practice that. And that's typically a year or so from when we put equipment in their data center, where they take their software, our equipment, put it together and practice and test on it so that they know they will never ever lose any data under any conditions. And that process takes 2 to 3 years. And depending on how aggressive the hyperscalers are, some of them are very aggressive. They move really quickly. Others are more methodical, and we're working with one of the more methodical ones now. And it's a 3-year effort, and they're not trying to shorten it up. They're taking their 3 years and getting it right, but they're long -- it's -- I guess the only expression is the bigger the plane, the longer the runway.

And I would add, Jamie, too. So there's some other consideration, including what I mentioned when I did my presentation in regards to environmental controls. So there's building considerations on their side, they need to have the locations where they've placed tape, right? That's sometimes a little bit a challenge for some of them. And then there's also usually with the big hyperscalers, we have an integrator in the middle. So you got to work with the integrators to be fully absorbed in their process also and those come back and forth in terms of qualification going on in that aspect as well.

Thanks, Bruno. The next question that came in is for Liz. Liz, you talked a lot about your -- the transformation of your team and your strategy. Can you just comment a little bit about how you're hiring and compensation practices have changed to embrace the shift to selling solutions as a service form factor?

Absolutely. So as I mentioned in my presentation, we have been strategically hiring to where actually the puck will be. And so we've brought in a lot of talent into our organization in a very strategic, well optimized way. So that we have -- we have experts in video surveillance. We've exports -- experts in autonomous driving. We have experts on software sales. We have experts in the enterprise, experts in healthcare. So we've really been very, very deliberate about where we want to go and are hiring for the future. As it relates to compensation, we actually redesigned the compensation plan last year when I joined. And then I actually modified it again this year so that we're driving a solutions-led evolution as I talked about solutions selling, you don't just sell a product. You sell a lot with it. You sell services, you lead with services. It's a bigger discussion and obviously, a bigger contract value as a result. So we put incentives into the compensation plan to really drive a more services-led discussion as well as also driving more of a margin-based focus as well as for our leaders in top line revenue focus. So it's a balance, but the whole idea is to drive a richer revenue and margin mix in everything we sell. And the sales organization is incented specifically to deliver that.

That's helpful. I have one for Ed. Ed, given your experience with Isilon, how would you compare StorNext versus Isilon on unstructured data?

Yes, they actually compete in very different spaces. If you look at -- Isilon is typically easy to use kind of set it and forget it. It's not designed to be fast. It was designed just to store data. StorNext's typically sits in front of Isilon where you want that blazing performance. So the biggest difference is speed, the cost of speed, very different for StorNext than it is for Isilon. Jamie alluded to another file system coming out. That other file system was more targeted at the Isilon type environments where easy to use, easy to manage, it takes you longer to rack it and stack it than it does to actually manage petabytes of data. And so that's -- I think what you'll see is that walking into an environment today, you'll see StorNext on the front, some other type of storage on the back like an Isilon will actually have an offering that will actually allow us to tier back and forth between the traditional StorNext and our own kind of Isilon in marketplace.

All right. And Mike, a couple of questions about the long-term model here. How should investors expect the service and maintenance revenue to trend over the next few years as the transition to software plays out?

Yes. I mean we'll have the continued headwinds with the backup tape. So we expect that to continue, but as Liz has outlined, services is a key part of the commission plan a part of the solution selling. So we would expect to be steady, if not increasing. And in particular, when we go to storage as a service, that will just augment that. So we really don't see a lot of disruption there, given how it's structured.

Great. And throughout the presentation, we didn't bring up royalties. We had a question from an investor. And just wanted to get sort of some general assumptions for royalties in the model going forward and your expectations for royalties going forward?

Well, when we look at royalties, we become less and less dependent on new royalty stream. In '20 -- in our fiscal year '20, it was a total of $20 million. When we look forward, it's probably going to be lower than that. So it's going to become less and less relevant to us. There will be probably a pickup when LTO-9 is adopted. So that will probably help it along, but it's an area that is great to have, but it's not one of our strategic focuses at this point.

Great. And some questions about the current environment? Obviously, COVID had a meaningful impact to M&E. What are the verticals that you've seen helping to offset the industry challenges? And just in general, are you seeing signs of recovery or dropping in the market?

All right. There is signs of recovery and there are signs of adaptation. I think if you look at, good examples are sports, the NBA, major league baseball and soon the NFL are resuming play but they're doing it in an adaptive way, right? It's not the way it used to be, but they've adapted to a new way of doing it. And so there are -- I'll be very honest, from my perspective, it feels like there are 2 economies. There is an economy of people who are accelerating in the pandemic, and there's an economy of people who are in trouble. Right? Amazon, Target, Walmart are taking off, Morgan Taylor, Macy's, Brooks Brothers are doomed. And so one of the things that we are doing is we're aligning with the winners, right? There's a lot of pressure on media and entertainment. There's a lot of pressure on traditional television, traditional movies, traditional sports, but as people are still getting entertained, but they're getting entertained on Instagram, TikTok, YouTube, Facebook. And so our media and entertainment team is chasing those dollars. It's very hard to make a TV show, but man, is it easy to post on Instagram. So we are realigning to the new economy. We're realigning to the winners, right? It's really tough for hotels. It's really tough for airports, but Airbnb is going like crazy, Etsy is going like crazy, Walmart is going like crazy. We are realigning to that winning economy, and we're going to wait for the traditional -- or the tougher part of the economy, easier to recover or not, but we just can't sit around and wait for them. We're just totally readjusting to the new economy, and we're designing our products, we're designing our sales team to go after that new economy.

And within that new economy, like where are the sectors that you're seeing the most meaningful opportunities right now?

Yes. A lot of it has to do with government suites and the fact that we're towards -- we're at an election year, government spending is strong, particularly in intelligence. I think given what's happening in the world, there's a lot of spending on in the intelligence community. A huge amount of spending in cloud. We're actually starting to see cloud opportunities that do not involve tape. We've been looking into other technologies we have, particularly in edge computing and very high-speed file systems. So we're seeing a lot of activity in cloud, research, university research and healthcare, medical, life sciences research is on fire. Online presences beyond cloud, again, the Etsy's, the Robinhood, the Schwab's online business, those online businesses that have been very resilient during COVID. Everyone is doubling down on those businesses. And you see that like Office Depot struggling in their stores were doing great online. They're just doubling down on those online businesses, and that's where we're aligning to those as well. So I think that -- those are the areas where we see a lot of green shoots, and we're aligning to that. And I -- the other thing we're seeing is our newer products are taking root. We've been chasing autonomous, had fits and spurts in autonomous. Now we're really in serious conversations in autonomous. We're getting much more traction than we've gotten before. Similarly, in surveillance, so I think we have businesses that are just taking root. And some of those are just new green shoots that are growing regardless of COVID.

And sort of those new areas, are they pulling you to the as-a-service model? I mean is it that -- their preferred procurement factor?

In many cases, it is, mainly because, I think, people have gotten used to it with the cloud. They like it. Like, look, I like paying for what I use. I don't necessarily know how much I'm going to use, so I like the ability to have an elastic consumption model. I really see most of the value in your software. So I like expressing my spending in terms of software because that's where our customers see the value versus just in-commodity servers and [iron]. So yes, I think we're getting pulled in this direction increasingly. And the other thing is, I think customers are starting to say, as they've integrated cloud more in their business, they're saying, let me do my business. If my business is sequencing genes, let me sequence genes and develop drugs, and I don't want a single person in our company thinking about storage. You do it. You give us the gear, you give us the software, you give us the monitoring, you give us the expertise. You patch it, you upgrade it, you keep it running. I just want to sequence genes like crazy, I want to study genes, and I just want to have Quantum give me bottomless storage, just never-ending storage to do my work and send me a bill. A reasonable bill, a bill that doesn't surprise me but you run all that for us and let us just do genes every which way, and you guys are the storage experts, why do I have to build an IT department with all these experts in Quantum. I never want to look at Quantum again. You just provide me the tag -- and provide me the tag where I need it, when I need it, and I want to just pay for a bill like I do for dial tone. And I think that's becoming, especially for the economies that are exploding because of COVID, their need for speed, speed is 4x money. Speed is so much more valuable than money. They're like Quantum help me go faster. Let me develop a drug faster. Let me get to Mars faster. Let me get my online business going faster, and I'm going to achieve that speed, if I just let you do all the, basically, storage as a service for us. And that we've really been pulled in that direction faster than we, I think, expected.

And sort of that need to speed and the customers' demand. As it relates to data and orchestration capabilities, I mean, do you have the offering that you need today? And what still needs to be developed? And is this organic? Or is this M&A? Sort of how do you think about that product roadmap?

Yes. I mean, Ed and I are collaborating really tightly on data orchestration and data -- think of it as a data catalog, a giant card catalog of everything you have is one tech and another tech is once you know your card catalog, everything you have, how do you move it around the world to the right place at the right time. Those are 2 different pieces of technology. One is the data catalog, the other is a data orchestrator. It's organizing and orchestrating and conducting movements all over the world. We've decided architecturally that both of those technologies have to sit in the cloud. Cloud is an overseer of all your locations, of all your data, and we are building some of that technology. We are acquiring some of that technology. Ed has really been a pioneer in both data catalogs and data orchestration. We've got pieces of that technology with Atavium but we're probably going to be doing a significant amount more work internally, looking at partners externally to build this because it's a very complex problem to solve. It will take us several years to solve it. We're not there today. We have pieces. And we've been pretty successful with those pieces. Those tend to be the sizzle on the steak that classification and orchestration technologies we have, the tiering we have, but I think we have a long road in front of us in that space, but we're committed that that's the architectural control point. That is the strategic control point of storage architectures and we're putting a lot of attention in that area.

Great. Jamie, you laid out this long-term plan today that you described as audacious. Wanted to get a sense of -- an investor asked, what are sort of the risk that keep you up at night in this transformation? And how you think about addressing those?

I'm mainly -- yes. I'm not worried about the strategy. I think we've vetted it, and our customers have told us we're bang on, on the strategy, Gartner, IDC, Fred, everyone is validating the strategy. So I'm feeling pretty good about the strategy. I'm feeling really good about operating the business day-to-day. I think we're running the business. We can always do better, but I feel like we're doing a pretty tight job of running the business. And all of my worry, all of my focus, all that keeps me up at night is velocity. Everything is coming down on velocity, right? We're not unique in our vision. Others see this opportunity and who's going to get there first. And really, my message to my teams internally, the message to my leadership team, the message to our salespeople and our engineers is we have got to move on this, and we have got to move fast. That is the biggest thing on my mind. When you're stuck in transition, I mean, I worked at a giant networking company that -- I mean, we were in transition for 15 years. I mean, it just -- we can't be in that kind of space. We have got to transition rapidly. We got to get to the other end of this thing rapidly, and we've got to get these new solutions we're talking about into our customers' hands faster than anyone else. So I'm literally getting kind of boring to my team because I'm like, at the state we're at, it's 10% strategy, 90% execution is where our heads are moving now.

Yes. And in terms of the velocity, I mean, are you seeing the early adopters engaging you now. Like where are they in terms of your offering and their ability to take it?

We started a new kind of dialogue with our customers. So a lot of our movie making customers just said, "Hey, guess what, our studios are moving to the cloud." And we ask them, well, so we go there together, so we build your cloud studio together, and they were like, yes, we love it. You guys are the perfect partner for that. So a lot of the transitions we're doing -- actually, every single transition we're doing, we're doing with a customer. We built the F-Series with a customer. They were -- our first big $3 million order came from our customers. Our cloud-based analytics, which is the foundation of all our services, we built with a customer. As now, working with a large media and entertainment company, we're building them a cloud studio, all based on StorNext. Right? We built our cloud offers with the biggest cloud company in the world. So we're building now, I think we're with the #1 and #2 automakers in the world, and we're helping them build autonomous vehicles, and we're learning together. So all these transitions we're doing with customers. And they are basically -- part of the velocity message is coming from them. Like, we got to get to the cloud. We've got to get these autonomous vehicles to market. We have got to get this done, and they're pushing us to make these transitions quicker. And I -- so everyone want -- people listen to me telling them to go fast, but I think they listen more when our customers are saying go fast. And I've been pretty pleased with our velocity. We got -- we're doing a lot. I will tell you, we're running more simultaneous programs, to say it's 3 to 4x more than we've done in the past is probably even conservative. We are running many programs in parallel and exercising a lot of new muscles about how to launch this many products, how to update the analyst and I mean the industry analyst community, Gartner, IDC and others in what we're doing. And as part of our velocity, we have to break out of the cubby we're in. Quantum is in a cubby. We've been put in a cubby, right? You're the tape guys, right? You're the old hard drive guys. And we're driving this message. We're driving the products with such intensity and such philosophy because we are just committed to break out of this cubby. We're not just old tape guys. We built the world's fastest file system. We help the world make movies, right? We help discover disease prevention through genomic analysis with our incredibly fast file systems. Right? We do some of the world's most advanced object storage that exists and cloud storage that exists or not the old tape type, but that's the reception, and we are committed to breaking out of that. And the way that I know to break out of that is velocity, just put out new products, win new customers, break new barriers, enter new markets with such velocity that we just decimate that cubby that people are putting us into. And again, for me, all roads lead to velocity for this company going forward.

Well, I think that's an important and probably good message to end on. We've given a lot today, and I know we had a target to be done at the hour. We're 6 minutes over. So Jamie, you want to maybe just make a closing statement?

Yes. We are taking on a lot, but I think the best strategies, the best companies have what are called BHAG, big hairy audacious goals. I think you need them. You need them to attract talent, you need them to attract investors, to attract customers. You've got to be doing something bold and exciting in our industry or you're roadkill. And I think we've got something pretty bold and exciting for us to do. And it's a multiyear mission. And I think great companies take on multiyear missions and I think great leaders don't think about their business in 90-day increments. We have to think about our business in half decade and decade-long increments to achieve something really great and that's what we put forward. It's the first time we've articulated something like this outside the company. We've been talking about it for almost 2 years and laying down the framework for this. It's still going to take some time. We haven't fixed everything in this company. We still have a scar tissue from things that happened to us years ago. Most of those wounds are healing, but we still have some of them. We still have areas that are not totally transformed, and -- but I think if you look at where we're starting at, from a market cap perspective, from a revenue perspective, there's just so much upside here, but we're just surrounded in upside. And I think -- I think it's going to be a lot of fun for the next 5 years. I'm signed up. Mike, I think you're signed up. I think everyone on this call was signed up for 5 years. We've got nothing better to do and nothing more interesting to do. So I think we're going to be talking with everyone on this call pretty frequently for years to come.

Well, Jamie, Mike and everybody, thank you for your time, for everybody that joined us today. Please note that the presentation that was used during today's event will be posted to the company's IR website, and we'll have an archived replay of this webcast on the site up there for you to review. And with that, we're going to conclude today's program, and thank you, everyone, for joining us.