(inspirational music) >> Hey, welcome back everybody. Jeff Frick here with theCUBE. We are in our Palo Alto studio. The conference season hasn't really kicked off yet into full swing so we can do a lot more kind of intimate stuff here in the studio, for a CUBE Conversation. And we're really excited to have a many time CUBE alum on, and a new guest, both from Western Digital. So Dave Tang, Senior Vice President at Western Digital. Great to see you again, Dave. >> Great to be here, Jeff. >> Absolutely and Martin Fink, he is the Chief Technology Officer at Western Digital, a longtime HP alum. I'm sure people recognized you from that and our great machine keynotes we were talking about it. So great to finally meet you, Martin. >> Thank you, nice to be here. >> Absolutely, so you guys are here talking about and we've got an ongoing program actually with Western Digital about Data Makes Possible, right. With all the things that are going on in tech at the end of the day, right, there's data, it's got to be stored somewhere and then of course there's processes and things going on. We've been exploring media and entertainment, sports, healthcare, autonomous vehicles, you know. All the places that this continues to reach out and it's such a fun project because you guys are a rising tide, lifts all boats, kind of company and really enjoy watching this whole ecosystem grow. So I really want to thank you for that. But now there's some new things that we want to talk about that you guys are doing to continue really in that same theme, and that's the support of this RISC-V. So first off, for people who have no idea, what is RISC-V? Let's jump into that and then kind of what is the announcement and why it's important. >> Sure, so RISC-V is an, you know, the tagline is, it's an open source instruction set architecture. So what does that mean, just so people can kind of understand. So today the world is dominated by two instruction set architectures. For the most part the, we'll call the desktop enterprise world is dominated by the Intel instruction set architecture and that's what's in most PCs, what people talk about as x86. And then the embedded and mobile space tends to be dominated by Arm, or by Arm Holdings. And so both of those are great architectures but they're also proprietary, they're owned by their respective companies. So RISC-V is essentially a third entrant, we'll say, into this world, but the distinction is that it's completely open source. So everything about the instruction set is available to all and anybody can implement it. We can all share the implementations. We can share the code that makes up that instruction set architecture, and very importantly for us and part of our motivation is the freedom to innovate. So we now have the ability to modify the instruction set or change the implementation of the instruction set, to optimize it for our devices and our storage and our drives, etc. >> So is this the first kind of open source play in microprocessor architecture? >> No, there's been other attempts at this. OpenSpark kind of comes to mind, and things like that, but the ability to get a community of individuals to kind of rally around this in a meaningful way has really been a challenge. And so I'd say that right now, RISC-V presents probably the best sort of clean slate, let's take some thing new to the market out there. >> So open source, obviously we've seen you know, take over the software world, first in the operating system which everybody is familiar with Linux but then we see it time and time again in different applications, Hadoop. I mean, there's just a proliferation of open source projects. The benefits are tremendous. Pretty easy to ascertain at a typical software case, how is that going to be applied do you think within the microprocessor world? >> So it's a little bit different. When we're talking about open source hardware or open source chips and microprocessors, you're dealing with a physical device. So even though you can open source all of the designs and the code associated with that device, you still have to fabricate it. You still have to create a physical design and you still have to call up a fab and say, will you make this for me at these particular volumes? And so that's the difference. So there are some differences between open source software where it's, you know, you create the bits and then you distribute those bits through the Internet and all is good. Whereas here, you still have a physical need to fabricate something. >> Now, how much more flexibility can you do then for the output when you can actually impact the architecture as opposed to just creating a custom chip design, on top of somebody else's architecture? >> Well, let me give you probably a really simple, concrete example that kind of people can internalize of some of our motivation behind this, because that might sort of help get people through this. If you think of a very typical surveillance application, you have a camera pointed into a room or a hallway. The reality is we're basically grabbing a ton of video frames but very few of them change, right? So the typical surveillance application is it never changes and you really want, only know when stuff changes. Well, today, in very simple terms, all of those frames get routed up to some big server somewhere and that server spends a lot of time trying to figure out, okay have I got a frame that changed? Have I got a frame that changed, and so on. And then eventually it'll find maybe two or three or five frames that have got something interesting. So in the world what we're trying to do is to say, okay well why don't we take that, find no changes, and push that right down to the device? So we basically store all those frames, why don't we go figure out all the frames that mean nothing, and only ship up to that big bad server the frames that have something interesting and something you want to go analyze and do some work on? So that's a very typical application that's quite meaningful because we can do all of that work at the device. We can eliminate shipping a whole bunch of data to where it's just going to get discarded anyways, and we can allow the end customer to really focus on the data that matters, and get some intelligence. >> And that's critical as we get more and more immersed in a data-centric world, where we have realtime applications like Martin described as well as large data-centric applications like of course, big data analytics, but also training for AI systems or machine learning. These workloads are going to become more and more diverse and they're going to need more specialized architectures and more specialized processing. So big data is getting bigger and faster and these realtime fast data applications are getting faster and bigger. So we need ways to contend with that, that really go beyond what's available with general purpose architectures. >> So that's a great point because if we take this example of video frames, now if I can build a processor that is customized to only do that, that's the only thing it does. It can be very low power, very efficient, and do that one thing very very well, and the cost adder, if you want to call it that, to the device where we put it, is a tiny fraction, but the cost savings of the overall solution is significant. So this ability to customize the instruction set to only do what you need it to do for that very special purpose, that's gold. >> So I just wanted to, Dave, we've talked about a lot of interesting innovations that you guys have come up with over the years, with the helium launch. Which I don't know, a couple, two, three years ago, you were just at the MAMR event, really energy assisted recording. So this is really kind of foundational within the storage and the media itself and how you guys do better and take advantage of evolving land space. This is a kind of a different play for Western Digital, this isn't a direct kind of improvement in the way that storage media and architecture works but this is really more of, I'm going to ask you. What is the Western Digital play here? Why is this an important space for you guys in your core storage business? >> Well we're really broadening our focus to really develop and innovate around technologies that really help the world extract more value from data as a whole, right. So it's way beyond storage these days, right. We're looking for better ways to capture, preserve, access, and transform the data. And unless you transform it, you can't really extract the value out of it so as we see all these new applications for data and the vast possibilities for data, we really want to pave the path and help the industry innovate to bring all those applications to reality. >> It's interesting too because one of the great topics always in computing is you know, you got compute and store, which has to go to which, right. And nobody wants to move a lot of data, that's hard and may or may not be easy to get compute. Especially these IoT applications, remote devices, tough conditions and power, which we mentioned a little bit before we went on air. So the the landscape for the for the need for compute and store in networking is radically changing than either the desktop or what we're seeing a consolidation in clouds. So what's interesting here, where does the scale come, right? At the end of the day, scale always wins. And that's where we've seen historically where the general-purpose microprocessor architectures is dominated but used to be a slew of specialty purpose architectures but now there's an opportunity to bring scale to this. So how does that scale game continue to evolve? >> So it's a great point that scale does matter and we've seen that repeatedly and so it's a significant part of the reason why we decided to go early with a significant commitment was to tell the world that we were bringing scale to the equation. And so what we communicated to the marketplace is we ship on the order of a billion processor cores a year, most people don't realize that all of our devices from USB sticks to hard drives, all have processors on them. And so we said, hey we're going to basically go all-in and go big and that translates into a billion cores that we ship every year and we're going to go on a program to essentially migrate all of those cores to RISC-V. It'll take a few years to get there but we'll migrate all of those cores and so we basically were signaling to the market, hey scale is now here. Scale is here, you can make the investments, you can go forward, you can make that commitment to RISC-V because essentially we've got your back. >> So just to make sure we get that clear. So you guys have announced that you're going to slowly migrate over time your micro processors that power your devices to the tune of approximately a billion with a B, cores per year to this new architecture. >> That is correct. >> And has that started? >> So the design has started. So we have started to design and develop our first two cores but the actual manifestation into devices probably in the early stage of 2020. >> Okay, okay. But that's a pretty significant commitment and again, the ideas you explicitly said it's a signal to the ecosystem, this is worth your investment because there is some scale here. >> Martin: That's right. >> Yeah, pretty exciting. And how do you think it's going to open up the ability for you to do new things with your devices that you before either couldn't do or we're too expensive with dollars or power. >> Martin: So we're going to step and iterate through this and one key point here is a lot of people tend to want to start in this processor world at the very high end, right. I'm going to go take on a Xeon processor or something like that. It's not what we're doing. We're basically saying, we're going to go at the small end, the tiny end where power matters. Power matters a lot in our devices and where can we achieve the optimum combination of power and performance. So even in our small devices like a USB stick or a client SSD or something like that, if we can reduce power consumption and even just maintain performance that's a huge win for our customers, you know. If you think about your laptop and if I reduce the power consumption of that SSD in there so that you have longer battery life and you can get you know through the day better, that's a huge win, right. And I don't impact performance in the process, that's a huge win. So what we do, what we're doing right now is we're developing the cores based on the RISC-V architecture and then what we're going to do is once we've got that sort of design, sort of complete is we want to take all of the typical client workloads and profile them on that. Then we want to find out, okay where are the hot spots? What are the two or three things that are really consuming all the power and how do we go optimize, by either creating two or three instructions or by optimizing the micro architecture for an existing instruction. And then iterate through that a few times so that we really get a big win, even at the very low end of the spectrum and then we just iterate through that with time. >> We're in a unique position I think in that the technologies that we develop span everything from the actual media where the bits are stored, whether it's solid-state flash or rotating magnetic disk and the recording heads. We take those technologies and build them all the way up into devices and platforms and full-fledged data center systems. And if we can optimize and tune all the way from that core media level all the way up through into the system level, we can deliver significantly higher value, we believe, to the marketplace. So this is the start of that, that enables us to customize command sets and optimize the flow of data so that we can we can allow users to access it when and where they need it. >> So I think there's another actually really cool point, which goes back to the open source nature of this and we try to be very clear about this. We're not going to develop our cores for all applications. We want the world to develop all sorts of different cores. And so for many applications somebody else might come in and say, hey we've got a really cool core. So one of the companies we've partnered with and invested in for example, is Esperanto. They've actually decided to go at the high end and do a machine learning accelerator. Hey, maybe we'll use that for some machine learning applications in our system level performance. So we don't have to do it all but we've got a common architecture across the portfolio and that speaks to that sort of open source nature of the RISC-V architecture is we want the world to get going. We want our competitors to get on board, we want partners, we want software providers, we want everybody on board. >> It's such a different ecosystem with open-source and the way the contributions are made and the way contributions are valued and the way that people can go find niches that are underserved. It's this really interesting kind of bifurcation of the market really, you don't really want to be in the big general-purpose middle anymore. That's not a great place to be, there's all kinds of specialty places where you can build the competence and with software and you know with, thank goodness for Moore's law decreasing prices of the power of the compute and now the cloud, which is basically always available. Really a exciting time to develop a myriad of different applications. >> Right and you talked before about scale in terms of points of implementation that will drive adoption and drive this to critical mass but there's another aspect of scale relative to the architecture within a single system that's also important that I think RISC-V helps to break down some barriers. Because with general purpose computer architectures, they assume a certain ratio of memory and storage and processing and bandwidth for interconnect and if you exceed those ratios, you have to add a whole new processor. Even though you don't need to need the processing capability, you need it for scale. So that's another great benefit of these new architectures is that the diversity of data needs where some are going to be large data sets, some are going to be small data sets that need need high bandwidth. You can customize and blend that recipe as you need to, you're not at the mercy of these fixed ratios. >> Yeah and I think you know it's so much of kind of what is cloud computing. And the atomic nature of it, that you can apply the ratios, the amount that you need as you need, you can change it on the fly, you can tone it up, tone it down. And I think the other interesting thing that you touched on is some of these new, which are now relatively special-purpose but are going to be general-purpose very soon in terms of machine learning and AI and applying those to different places and applying them closer to the problem. It's a very very interesting evolution of the landscape but what I want to do is kind of close on you Martin, especially because again kind of back to the machine. Not the machine specifically but you have been in the business of looking way down the road for a long time. So you came out, I'd looked at your LinkedIn, you retired for three months, congratulations. (laughs) Hope you got some my golf in but you came back to Western Digital so why did you come back? And as you look down the road a ways, what do you see that it excites you, that got you off that three-month little tour around the golf course and I'm sorry I had to tease about that. But what do you see? What are you excited about that you came back and got involved in an open source microprocessor project? >> So the the short answer was that, I saw the opportunity at Western Digital to be where data lives. So I had spent my entire career, will call it at the compute or the server side of things and the interesting thing is I had a very close relationship with SanDisk, which was acquired by Western Digital. And so I had, we'll call it an insider view, of what was possible there and so what triggered was essentially what we're talking about here was given that about half the world's data lands on Western Digital devices, taking that from a real position of strength in the marketplace and say, what could we go do to make data more intelligent and rather than start kind of at that server end and so that I saw that potential there and it was just incredible, so that's that's what made me want to join. >> Exciting times. Dave good get. (laughs) >> We're delighted to have Martin with us. >> All right, well we look forward to watch it evolve. We've got another another whole set of events we're going to do again together with Western Digital that we're excited about. Again, covering Data Makes Possible but you know kind of uplifting into the application space as a lot of the cool things that people are doing in innovation. So Martin, great to finally meet you and thanks for stopping by. >> Thanks for the time. >> David as always and I think we'll see in a month or so. >> Right, always a pleasure Jeff, thanks. >> All right Martin Fink, Dave Tang. I'm Jeff Frick, you're watching theCUBE. Thanks for watching, we'll catch you next time. (inspirational music)

>> Announcer: Live from San Jose, California, it's theCUBE covering innovating to fuel the next decade of big data. Brought to you by Western Digital. >> Hey, welcome back everybody. Jeff Frick, here with theCUBE. We are at the Western Digital Headquarters in San Jose, California. It's the Al-Mady Campus. A historic campus. It's had a lot of great innovation, especially in hard drives for years and years and years. This event's called Innovating to Fuel the Next Data Big Data. And we're excited to have a big brain on. We like to get smart people who's been watching this story for a while and will give us a little bit of historical perspective. It's John Rydning. He is the Research Vice President for Hard Drives for IEC. John, Welcome. >> Thank you, Jeff. >> Absolutely. So, what is your take on today's announcement? >> I think it's our very meaningful announcement, especially when you consider that the previous BIGIT Technology announcement for the industry was Helium, about four or five years ago. But, really, the last big technology announcement prior to that was back in 2005, 2006, when the industry announced making this transition to what they called at that time, "Perpendicular Magnetic Recording." And when that was announced it was kind of a similar problem at that time in the industry that we have today, where the industry was just having a difficult time putting more data on each disc inside that drive. And, so they kind of hit this technology wall. And they announced Perpendicular Magnetic Recording and it really put them on a new S curve in terms of their ability to pack more data on each disc and just kind of put it in some perspective. So, after they announce Perpendicular Magnetic Recording, the capacity per disc increased about 30% a year for about five years. And then over, really, a ten year period, increased about an average of about 20% a year. And, so today's announcement is I see a lot of parallels to that. You know, back when Perpendicular Magnetic Recording was announced, really they build. They increased the capacity per platter was growing very slowly. That's where we are today. And with this announcement of MAMR Technology the direction that Western Digital's choosing really could put the industry on a new S curve and putting in terms of putting more capacity, storage capacity on each one of those discs. >> It's interesting. Always reminds me kind of back to the OS in Microsoft in Intel battles. Right? Intel would come out with a new chip and then Microsoft would make a bigger OS and they go back and back and forth and back and forth. >> John: Yeah, that's very >> And we're seeing that here, right? Cuz the demands for the data are growing exponentially. I think one of the numbers that was thrown out earlier today that the data thrown off by people and the data thrown off by machines is so exponentially larger than the data thrown off by business, which has been kind of the big driver of IT spin. And it's really changing. >> It's a huge fundamental shift. It really is >> They had to do something. Right? >> Yeah, the demand for a storage capacity by these large data centers is just phenomenal and yet at the same time, they don't want to just keep building new data center buildings. And putting more and more racks. They want to put more storage density in that footprint inside that building. So, that's what's really pushing the demand for these higher capacity storage devices. They want to really increase the storage capacity per cubic meter. >> Right, right. >> Inside these data centers. >> It's also just fascinating that our expectation is that they're going to somehow pull it off, right? Our expectation that Moore's laws continue, things are going to get better, faster, cheaper, and bigger. But, back in the back room, somebody's actually got to figure out how to do it. And as you said, we hit these kind of seminal moments where >> Yeah, that's right. >> You do get on a new S curve, and without that it does flatten out over time. >> You know, what's interesting though, Jeff, is really about the time that Perpendicular Magnetic Recording was announced way back in 2005, 2006, the industry was really, already at that time, talking about these thermal assist technologies like MAMR that Western Digital announced today. And it's always been a little bit of a question for those folks that are either in the industry or watching the industry, like IDC. And maybe even even more importantly for some of the HDD industry customers. They're kind of wondering, so what's really going to be the next technology race horse that takes us to that next capacity point? And it's always been a bit of a horse race between HAMR and MAMR. And there's been this lack of clarity or kind of a huge question mark hanging over the industry about which one is it going to be. And Western Digital certainly put a stake in the ground today that they see MAMR as that next technology for the future. >> (mumbles words) Just read a quote today (rushes through name) key alumni just took a new job. And he's got a pin tweet at the top of his thing. And he says, "The smart man looks for ways "To solve the problem. "Or looks at new solutions. "The wise man really spends his time studying the problem." >> I like that. >> And it's really interesting here cuz it seems kind of obvious there. Heat's never necessarily a good thing with electronics and data centers as you mentioned trying to get efficiency up. There's pressure as these things have become huge, energy consumption machines. That said, they're relatively efficient, based on other means that we've been doing they compute and the demand for this compute continues to increase, increase, increase, increase. >> Absolutely >> So, as you kind of look forward, is there anything kind of? Any gems in the numbers that maybe those of us at a layman level are kind of a first read are missing that we should really be paying attention that give us a little bit of a clue of what the feature looks like? >> Well, there's a couple of major trends going on. One is that, at least for the hard drive industry, if you kind of look back the last ten years or so, a pretty significant percentage of the revenue that they've generated a pretty good percentage of the petabytes that they ship have really gone into the PC market. And that's fundamentally shifting. And, so now it's really the data centers, so that by the time you get to 2020, 2021, about 60 plus percent of the petabytes that the industry's shipping is going into data centers, where if you look back a few years ago, 60% was going into PCs. That's a big, big change for the industry. And it's really that kind of change that's pushing the need for these higher capacity hard drives. >> Jeff: Right. >> So, that's, I think, one of the biggest shifts has taking place. >> Well, the other thing that's interesting in that comment because we know scale drives innovation better than anything and clearly Intel microprocessors rode the PC boom to get out scale to drive the innovation. And, so if you're saying, now, that the biggest scale is happening in the data center Then, that's a tremendous force for innovation in there versus Flash, which is really piggy-backing on the growth of these jobs, because that's where it's getting it's scale. So, when you look at kind of the Flash hard drive comparison, right? Obviously, Flash is the shiny new toy getting a lot of buzz over the last couple years. Western Digital has a play across the portfolio, but the announcement earlier today said, you're still going to have like this TenX cost differentiation. >> Yeah, that's right. >> Even through, I think it was 20, 25. I don't want to say what the numbers were. Over a long period of time. You see that kind of continuing DC&E kind of conflict between those two? Or is there a pretty clear stratification between what's going to go into Flash systems, or what's going to hard drives? >> That's a great question, now. So, even in the very large HyperScale data centers and we definitely see where Flash and hard disk drives are very complimentary. They're really addressing different challenges, different problems, and so I think one of the charts that we saw today at the briefing really is something that we agree with strongly at IDC. Today, maybe, about 7% or 8% of all of the combined HDD SSD petabyte shipped for enterprise are SSD petabytes. And then, that grows to maybe ten. >> What was it? Like 7% you said? >> 6% to 7%. >> 6% to 7% okay. Yeah, so we still have 92, 93%, 94% of all petabytes that again are HDD SSD petabytes for enterprise. Those are still HDD petabytes. And even when you get out to 2020, 2021, again, still bought 90%. We agree with what Western Digital talked about today. About 90% of the combined HDD SSD petabytes that are shipping for enterprise continue to be HDD. So, we do see the two technologies very complementary. Talked about SSD is kind of getting their scale on PCs and that's true. They really are going to quickly continue to become a bigger slice of the storage devices attached to new PCs. But, in the data center you really need that bulk storage capacity, low cost capacity. And that's where we see that the two SSDs and HDDs are going to live together for a long time. >> Yeah, and as we said the conflict barrier, complimentary nature of the two different applications are very different. You need the big data to build the models, to run the algorithms, to do stuff. But, at the same time, you need the fast data that's coming in. You need the real time analytics to make modifications to the algorithms and learn from the algorithms >> That's right, yeah. It's the two of those things together that are one plus one makes three type of solution. Exactly, and especially to address latency. Everybody wants their data fast. When you type something into Google, you want your response right away. And that's where SSDs really come into play, but when you do deep searches, you're looking through a lot of data that has been collected over years and a lot of that's probably sitting on hard disc drives. >> Yeah. The last piece of the puzzle, I just want to you to address before we sign off, That was an interesting point is that not just necessarily the technology story, but the ecosystem story. And I thought that was really kind of, I thought, the most interesting part of the MAMR announcement was that it fits in the same form factor, there's no change to OS, there's no kind of change in the ecosystem components in which you plug this in. >> Yeah, that's right. It's just you take out the smaller drive, the 10, or the 12, or whatever, or 14 I guess is coming up. And plug in. They showed a picture of a 40 terabyte drive. >> Right. >> You know, that's the other part of the story that maybe doesn't get as much play as it should. You're playing in an ecosystem. You can't just come up with this completely, kind of independent, radical, new thing, unless it'S so radical that people are willing to swap out their existing infrastructure. >> I completely agree. It's can be very difficult for the customer to figure out how to adopt some of these new technologies and actually, the hard disk drive industry has thrown a couple of technologies at their customers over the past five, six years, that have been a little challenging for them to adopt. So, one was when the industry went from a native 512 by sectors to 4K sectors. Seems like a pretty small change that you're making inside the drive, but it actually presented some big challenges for some of the enterprise customers. And even the single magnetic recording technologies. So, it has a way to get more data on the disc, and Western Digital certainly talked about that today. But, for the customer trying to plug and play that into a system and SMR technology actually created some real challenges for them to figure out how to adopt that. So, I agree that what was shown today about the MAMR technology is definitely a plug and play. >> Alright, we'll give you the last word as people are driving away today from the headquarters. They got a bumper sticker as to why this is so important. What's it say on the bumper sticker about MAMR? It says that we continue to get more capacity at a lower cost. >> (chuckles) Isn't that just always the goal? >> I agree. >> (chuckles) Alright, well thank you for stopping by and sharing your insight. Really appreciate it. >> Thanks, Jeff. >> Alright. Jeff Frick here at Western Digital. You're watching theCUBE! Thanks for watching. (futuristic beat)

>> Announcer: Live from San Jose, California, it's theCUBE, covering Innovating to Fuel the Next Decade of Big Data, brought to you by Western Digital. >> Hey welcome back everybody, Jeff Frick here with theCUBE. We're at Western Digital at their global headquarters in San Jose, California, it's the Almaden campus. This campus has a long history of innovation, and we're excited to be here, and probably have the smartest person in the building, if not the county, area code and zip code. I love to embarrass here, Janet George, she is the Fellow and Chief Data Scientist for Western Digital. We saw you at Women in Data Science, you were just at Grace Hopper, you're everywhere and get to get a chance to sit down again. >> Thank you Jeff, I appreciate it very much. >> So as a data scientist, today's announcement about MAMR, how does that make you feel, why is this exciting, how is this going to make you be more successful in your job and more importantly, the areas in which you study? >> So today's announcement is actually a breakthrough announcement, both in the field of machine learning and AI, because we've been on this data journey, and we have been very selectively storing data on our storage devices, and the selection is actually coming from the preconstructed queries that we do with business data, and now we no longer have to preconstruct these queries. We can store the data at scale in raw form. We don't even have to worry about the format or the schema of the data. We can look at the schema dynamically as the data grows within the storage and within the applications. >> Right, cause there's been two things, right. Before data was bad 'cause it was expensive to store >> Yes. >> Now suddenly we want to store it 'cause we know data is good, but even then, it still can be expensive, but you know, we've got this concept of data lakes and data swamps and data all kind of oceans, pick your favorite metaphor, but we want the data 'cause we're not really sure what we're going to do with it, and I think what's interesting that you said earlier today, is it was schema on write, then we evolved to schema on read, which was all the rage at Hadoop Summit a couple years ago, but you're talking about the whole next generation, which is an evolving dynamic schema >> Exactly. >> Based whatever happens to drive that query at the time. >> Exactly, exactly. So as we go through this journey, we are now getting independent of schema, we are decoupled from schema, and what we are finding out is we can capture data at its raw form, and we can do the learning at the raw form without human interference, in terms of transformation of the data and assigning a schema to that data. We got to understand the fidelity of the data, but we can train at scale from that data. So with massive amounts of training, the models already know to train itself from raw data. So now we are only talking about incremental learning, as the train model goes out into the field in production, and actually performs, now we are talking about how does the model learn, and this is where fast data plays a very big role. >> So that's interesting, 'cause you talked about that also earlier in your part of the presentation, kind of the fast data versus big data, which kind of maps the flash versus hard drive, and the two are not, it's not either or, but it's really both, because within the storage of the big data, you build the base foundations of the models, and then you can adapt, learn and grow, change with the fast data, with the streaming data on the front end, >> Exactly >> It's a whole new world. >> Exactly, so the fast data actually helps us after the training phase, right, and these are evolving architectures. This is part of your journey. As you come through the big data journey you experience this. But for fast data, what we are seeing is, these architectures like Lambda and Kappa are evolving, and especially the Lambda architecture is very interesting, because it allows for batch processing of historical data, and then it allows for what we call a high latency layer or a speed layer, where this data can then be promoted up the stack for serving purposes. And then Kappa architecture's where the data is being streamed near real time, bounded and unbounded streams of data. So this is again very important when we build machine learning and AI applications, because evolution is happening on the fly, learning is happening on the fly. Also, if you think about the learning, we are mimicking more and more on how humans learn. We don't really learn with very large chunks of data all at once, right? That's important for initially model training and model learning, but on a regular basis, we are learning with small chunks of data that are streamed to us near real time. >> Right, learning on the Delta. >> Learning on the Delta. >> So what is the bound versus the unbound? Unpack that a little bit. What does that mean? >> So what is bounded is basically saying, hey we are going to get certain amounts of data, so you're sizing the data for example. Unbounded is infinite streams of data coming to you. And so if your architecture can absorb infinite streams of data, like for example, the sensors constantly transmitting data to you, right? At that point you're not worried about whether you can store that data, you're simply worried about the fidelity of that data. But bounded would be saying, I'm going to send the data in chunks. You could also do bounded where you basically say, I'm going to pre-process the data a little bit just to see if the data's healthy, or if there is signal in the data. You don't want to find that out later as you're training, right? You're trying to figure that out up front. >> But it's funny, everything is ultimately bounded, it just depends on how you define the unit of time, right, 'cause you take it down to infinite zero, everything is frozen. But I love the example of the autonomous cars. We were at the event with, just talking about navigation just for autonomous cars. Goldman Sachs says it's going to be a seven billion dollar industry, and the great example that you used of the two systems working well together, 'cause is it the car centers or is it the map? >> Janet: That's right. >> And he says, well you know, you want to use the map, and the data from the map as much as you can to set the stage for the car driving down the road to give it some level of intelligence, but if today we happen to be paving lane number two on 101, and there's cones, now it's the real time data that's going to train the system. But the two have to work together, and the two are not autonomous and really can't work independent of each other. >> Yes. >> Pretty interesting. >> It makes perfect sense, right. And why it makes perfect sense is because first the autonomous cars have to learn to drive. Then the autonomous cars have to become an experienced driver. And the experience cannot be learned. It comes on the road. So one of the things I was watching was how insurance companies were doing testing on these cars, and they had a human, a human driving a car, and then an autonomous car. And the autonomous car, with the sensors, were predicting the behavior, every permutation and combination of how a bicycle would react to that car. It was almost predicting what the human on the bicycle would do, like jump in front of the car, and it got it right 80% of the cases. But a human driving a car, we're not sure how the bicycle is going to perform. We don't have peripheral vision, and we can't predict how the bicycle is going to perform, so we get it wrong. Now, we can't transmit that knowledge. If I'm a driver and I just encountered a bicycle, I can't transmit that knowledge to you. But a driverless car can learn, it can predict the behavior of the bicycle, and then it can transfer that information to a fleet of cars. So it's very powerful in where the learning can scale. >> Such a big part of the autonomous vehicle story that most people don't understand, that not only is the car driving down the road, but it's constantly measuring and modeling everything that's happening around it, including bikes, including pedestrians, including everything else, and whether it gets in a crash or not, it's still gathering that data and building the model and advancing the models, and I think that's, you know, people just don't talk about that enough. I want follow up on another topic. So we were both at Grace Hopper last week, which is a phenomenal experience, if you haven't been, go. Ill just leave it at that. But Dr. Fei-Fei Li gave one of the keynotes, and she made a really deep statement at the end of her keynote, and we were both talking about it before we turned the cameras on, which is, there's no question that AI is going to change the world, and it's changing the world today. The real question is, who are the people that are going to build the algorithms that train the AI? So you sit in your position here, with the power, both in the data and the tools and the compute that are available today, and this brand new world of AI and ML. How do you think about that? How does that make you feel about the opportunity to define the systems that drive the cars, et cetera. >> I think not just the diversity in data, but the diversity in the representation of that data are equally powerful. We need both. Because we cannot tackle diverse data, diverse experiences with only a single representation. We need multiple representation to be able to tackle that data. And this is how we will overcome bias of every sort. So it's not the question of who is going to build the AI models, it is a question of who is going to build the models, but not the question of will the AI models be built, because the AI models are already being built, but some of the models have biases into it from any kind of lack of representation. Like who's building the model, right? So I think it's very important. I think we have a powerful moment in history to change that, to make real impact. >> Because the trick is we all have bias. You can't do anything about it. We grew up in the world in which we grew up, we saw what we saw, we went to our schools, we had our family relationships et cetera. So everyone is locked into who they are. That's not the problem. The problem is the acceptance of bring in some other, (chuckles) and the combination will provide better outcomes, it's a proven scientific fact. >> I very much agree with that. I also think that having the freedom, having the choice to hear another person's conditioning, another person's experiences is very powerful, because that enriches our own experiences. Even if we are constrained, even if we are like that storage that has been structured and processed, we know that there's this other storage, and we can figure out how to get the freedom between the two point of views, right? And we have the freedom to choose. So that's very, very powerful, just having that freedom. >> So as we get ready to turn the calendar on 2017, which is hard to imagine it's true, it is. You look to 2018, what are some of your personal and professional priorities, what are you looking forward to, what are you working on, what's top of mind for Janet George? >> So right now I'm thinking about genetic algorithms, genetic machine learning algorithms. This has been around for a while, but I'll tell you where the power of genetic algorithms is, especially when you're creating powerful new technology memory cell. So when you start out trying to create a new technology memory cell, you have materials, material deformations, you have process, you have hundred permutation combination, and the genetic algorithms, we can quickly assign a cause function, and we can kill all the survival of the fittest, all that won't fit we can kill, arriving to the fastest, quickest new technology node, and then from there, we can scale that in mass production. So we can use these survival of the fittest mechanisms that evolution has used for a long period of time. So this is biology inspired. And using a cause function we can figure out how to get the best of every process, every technology, all the coupling effects, all the master effects of introducing a program voltage on a particular cell, reducing the program voltage on a particular cell, resetting and setting, and the neighboring effects, we can pull all that together, so 600, 700 permutation combination that we've been struggling on and not trying to figure out how to quickly narrow down to that perfect cell, which is the new technology node that we can then scale out into tens of millions of vehicles, right? >> Right, you're going to have to >> Getting to that spot. >> You're going to have to get me on the whiteboard on that one, Janet. That is amazing. Smart lady. >> Thank you. >> Thanks for taking a few minutes out of your time. Always great to catch up, and it was terrific to see you at Grace Hopper as well. >> Thank you, I really appreciate it, I appreciate it very much. >> All right, Janet George, I'm Jeff Frick. You are watching theCUBE. We're at Western Digital headquarters at Innovating to Fuel the Next Generation of Big Data. Thanks for watching.

>> Announcer: Live from San Jose, California, it's theCUBE, covering Innovating to Fuel the Next Decade of Big Data, brought to you by Western Digital. >> Hey welcome back everybody, Jeff Frick here with theCUBE. We're at Western Digital's headquarters in San Jose, California at the Almaden campus. Lot of innovation's been going on here, especially in storage for decades, and we're excited to be at this special press and analyst event that Western Digital put on today to announce some exciting new products. It's called Innovating to Fuel the Next Decade of Data. I'm super happy to have a long-time industry veteran, he just told me, 35 years, I don't know if I can tell (Mark laughs) that on air or not. He's Mark Grace, he's the Senior Vice President of Devices for Western Digital, Mar, great to have you on. >> Thanks Jeff, glad to be here. >> Absolutely, so you've seen this movie over and over and over, I mean that's one of the cool things about being in the Valley, is this relentless pace of innovation. So how does today's announcement stack up as you kind of look at this versus kind of where we've come from? >> Oh I think this is maybe one of the, as big as it comes, Jeff, to be honest. I think we've plotted a course now that I think was relatively uncertain for the hard drive industry and the data center, and plotted a course that I think we can speak clearly to the market, and clearly to customers about the value proposition for rotating magnetic storage for decades to come. >> Which is pretty interesting, 'cause, you know, rotating drives have been taking a hit over the last couple of years, right, flash has been kind of the sexy new kid on the block, so this is something new, >> Mark: It is. >> And a new S curve I think as John said. >> I agree, we're jumping onto a, we're extending the S curve, let's call it that. I think there's actually plenty of other S curve opportunities for us, but in this case, I think the industry, and I would say our customer base, we have been less than clear with those guys about how we see the future of rotating storage in the cloud and enterprise space, and I think today's announcement clarifies that and gives some confidence about architectural decisions relative to rotating storage going forward for a long time. >> Well I think it's pretty interesting, 'cause compared to the other technology that was highlighted, the other option, the HAMR versus the MAMR, this was a much more elegant, simpler way to add this new S curve into an existing ecosystem. >> You know, elegant's probably a good word for it, and it's always the best solution I would say. HAMR's been a push for many years. I can't remember the first time I heard about HAMR. It's still something we're going to continue to explore and invest in, but it has numerous hurdles compared to the simplicity and elegance, as you say, of MAMR, not the least of which is we're going to operate at normal ambient temperatures versus apply tremendous heat to try and energize the recording and the technologies. So any time you introduce extraordinary heat you face all kinds of ancillary engineering challenges, and this simplifies those challenges down to one critical innovation, which is the oscillator. >> Pretty interesting, 'cause it seems pretty obvious that heat's never a good thing. So it's curious that in the quest for this next S curve that the HAMR path was pursued for as long as it was, it sounds like, because it sounds like that's a pretty tough thing to overcome. >> Yeah, I think it initially presented perhaps the most longevity perhaps in early exploration days. I would say that HAMR has certainly received the most press as far as trying to assert it as the extending recording technology for enterprise HDDs. I would say we've invested for almost as long in MAMR, but we've been extremely quiet about it. This is kind of our nature. When we're ready to talk about something, you can kind of be sure we're ready to go with it, and ready to think about productization. So we're quite confident in what we're doing. >> But I'm curious from your perspective, having been in the business a long time, you know, we who are not directly building these magical machines, just now have come to expect that Moore's Law will contain, has zero to do with semiconductor physics anymore, it's really an attitude and this relentless pace of innovation that now is expected and taken for granted. You're on the other side, and have to face real physics and mechanical limitations of the media and the science and everything else. So is that something that gets you up every day >> Mark: Keeps me awake every night! >> Obviously keeps you awake at night and up every day. You've been doing it for 35 years, so there must be some appeal. >> Yeah. (laughs) >> But you know, it's a unique challenge, 'cause at the same time not only has it got to be better and faster and bigger, it's got to be cheaper, and it has been. So when you look at that, how does that kind of motivate you, the teams here, to deliver on that promise? >> Yeah, I mean in this case, we are a little bit defensive, in the sense of the flash expectations that you mentioned, and I think as we digest our news today, we'll be level setting a little bit more in a more balanced way the expectations for contribution from rotating magnetic storage and solid state storage to what I think is a more accurate picture of its future going forward in the enterprise and hyperscale space. To your point about just relentless innovation, a few of us were talking the other day in advance of this announcement that this MAMR adventure feels like the early days of PMR, perpendicular, the current recording technology. It feels like we understand a certain amount of distance ahead of us, and that's about this four-terabit per inch kind of distance, but it feels like the early days where we could only see so far but the road actually goes much further, and we're going to find more and more ways to extend this technology, and keep that order of magnitude cost advantage going from a hard drive standpoint versus flash. >> I wonder how this period compares to that, just to continue, in terms of on the demand side, 'cause you know, back in the day, the demand and the applications for these magical compute machines weren't near, I would presume, as pervasive as now, or am I missing the boat? 'Cause now clearly there is no shortage of demand for storage and compute. >> Yeah, depending on where you're coming from, you could take two different views of that. The engine that drove the scale of the hard drive industry to date has, a big piece of it in the long history of the hard drive industry has been the PC space. So you see that industry converting to flash and solid state storage more aggressively, and we embrace that, you know we're invested in flash and we have great products in that space, and we see that happening. The opportunity in the hyperscale and cloud space is we're only at the tip of the iceberg, and therefore I think, as we think about this generation, we think about it differently than those opportunities in terms of breadth of applications, PCs, and all that kind of create the foundation for the hard drive, but what we see here is the virtuous cycle of more storage, more economical storage begets more value proposition, more opportunities to integrate more data, more data collection, more storage. And that virtuous cycle seems to me that we're just getting started. So long live data, that's what we say. (both laugh) >> The other piece that I find interesting is before the PCs were the driver of scale relative to an enterprise data center, but with the hyperscale guys and the proliferation of cloud and actually the growth of PCs is slowing down dramatically, that it's kind of flipped the bit. Now the data centers themselves have the scale to drive >> Absolutely. >> the scale innovation that before was before was really limited to either a PC or a phone or some more consumer device. >> Absolutely the case. When you take that cross-section of hard drive applications, that's a hundred percent the case, and in fact, we look at the utilization as a vertically integrated company we look at our media facilities for the disks, we look at our wafer facilities for heads, and those facilities as we look forward are going to be as busy as busier than they've ever been. I mean the amount of data is relative to the density as well as disks and heads and how many you can employ. So we see this in terms of fundamental technology and component construction, manufacturing, busier than it's ever been. We'll make fewer units. I mean there will be fewer units as they become bigger and denser for this application space, but the fundamental consumption of magnetic recording technology and components is all-time records. >> Right. And you haven't even talked about the software-defined piece that's dragging the utilization of that data across multiple applications. >> And it's just one of these that come in to help everybody there too, yeah. >> Jeff: You got another 35 years more years in you? (both laugh) >> I hope so. >> All right. >> But that would be the edge of it, I think. >> All right, we're going to take Mark Grace here, only 35 more years, Lord knows what he'll be working on. Well Mark, thanks for taking a few minutes and answering your prospective >> No that's fine, thanks a lot. >> Absolutely, Mark Grace, I'm Jeff Frick, you're watching theCUBE from Western Digital headquarters in San Jose, California. Thanks for watching. >> Mark: All right.

>> Male voiceover: Live from San Jose California, it's the Cube, covering Innovating to Fuel the Next Decade of Big Data. Brought to you by Western Digital. >> Hey, welcome back everybody, Jeff Frick here with the Cube. We're at the Western Digital World Headquarters It's the Almaden Campus in San Jose. If you know anything about the tech world, you know there's a lot of innovation that's been happening on this campus for years and years and years. Big announcement today called Innovating to Fuel the Next Generation of Big Data. Lot of exciting announcements and here to join us to tell us all about it is Brendan Collins. He's the Vice President of Product Marketing Devices for Western Digital. Brendan, great to see you. >> Thank you, glad to be here. >> Absolutely so, really exciting announcement. You know, I've talked to Kim Stevenson at Intel, we had an interview talking about Moore's law. And one thing she really reinforced is that Moore's law is really more of an attitude than it is specifically physics, and whether you want to argue the physics is one thing, but the attitude for innovation, to continue to deliver a lot more for less, just continues, continues, and continues, and you guys announced a huge step in that direction today. >> Yeah, we have a challenge that storage is growing at a rate of about 40 percent per year. And budgets from the data centers are not growing, right? So the challenge is for us to develop new technologies that allow us to stay on the technology curve, and cut costs and do that efficiently. >> Then this is a big one, so let's jump in. So actually it was years ago I was actually at the event when you guys introduced the Helium drives, and that was a big deal there, and you've continued to kind of move that innovation but then you can see a plateau. And the density of this data, so you guys had to come up with something new. >> Yeah, what we've seen is that our PMR technology that we use currently is slowly running out of steam, right? So in order to come down the cost curve, we needed to boost areal density. And luckily we were able to come up with a new breakthrough in MAMR technology that will allow us to do that for the next decade. >> It's interesting in the talk, you talked about you guys could see this kind of coming and you actually put a lot of bets on the table, you didn't just bet on MAMR, you bet on HAMR, and you continued along a number of multiple tracks, and you've been at this for a while. What was kind of the innovation that finally gave you a breakthrough moment that got us to where we are today? >> Well, there were multiple technologies that we could have invested in, and we decided to continue on the two major ones which were HAMR and MAMR but we made a decision to invest in a process called, a head fabrication process called damascene that allowed us to extend the life of PMR for the last five to six years, and it's been in all the products we've been shipping since 2013. >> And you talked the areal density, so that's basically the amount of information we can put on the square inch of surface area And you've really, you attacked it on two vectors. One is how many tracks, just think of a record, how many tracks can you get on an album, in terms of the number of lines, and then how much density then you can have on each of those tracks. >> That's right, that's right. And you're now seeing major improvements on both of those factors. >> Well if you look at, we've had three enabling technologies in our products for the past three to four years, right. One is helium, one is micro actuation, and the other is the damascene process. Damascene and micro actuation actually push track density which enables higher capacity. But the newer technology that we're talking about, MAMR, addresses both factors. So we push the track density even tighter together, But we also boost the linear density at the same time, and we do that without adding cost. >> Right. The other thing you talked about, and I think it's a really important piece, right it's not only the technology breakthrough, but it's also how does that fit within the existing ecosystem of your customers, and obviously big giant data centers and big giant cloud providers, we actually have a show going on at a big cloud show right now, and this technology was innovative in that you've got a breakthrough on density, but not so crazy that you introduced a whole bunch of new factors into the ecosystem that would then have to be incorporated into all these systems, because you guys not only make your own systems, but you make the media that feeds a whole host of ecosystems, and that was a pretty important piece. >> If you look at some previous technologies we've introduced whether it be even 4K sectors in the industry, or shingled magnetic reporting, both of those require whole side modifications. Any time you have whole side modifications, it generally slows down the adoption, right? With HAMR, one of the challenges that we had was because of the concerns with thermals on the media, we needed a process called wear leveling, and that required whole software changes. In contrast, when we go to MAMR, everything is seamless, everything is transparent, and it's great. >> Right. I thought it was much simpler than that. I thought just heat is bad, HAMR is heat, and MAMR is microwave, and you know, heat and efficiency and data centers and all those, kind of again, system-level concerns; heat's never a good thing in electronics. >> Well, and in the case of MAMR versus HAMR, there's like an order of magnitude difference in the temperature on the disk, which is the key concern. >> And then of course as you mentioned in the key note, this is real, you've got sample units going on, correct me if I'm wrong, as early as next year >> That's right. >> you're hoping you'd be in scale production in 2020. Where some of these other competing technologies, there's really still no forecasted ship date on the horizon. >> Yeah, you can generate samples, you can build lower quantities of these HAMR drives, but you still have that big concern out there in front of you, how do I address the reliability, how do I address the complexity of all these new materials, and then if I got all of that to work, how do I do it commercially because of the cost additives. >> Right; so I just want to get your perspective before we let you go, you're busy, there's a high demand for your time, as you kind of think back and look at these increasing demands for storage, this increasing demand for computers, and I think one of the data points given is, you know, the data required for humans and machines and IOT is growing way way way way faster than business focused data which has been the driver of a lot of this stuff, if you just kind of sit back and take a look, you know, what are some of your thoughts because I'm sure not that long ago you could have never imagined that there would be the demand for the types of capacities that we're talking about now and we both know that when we sit down five years from now, ten years ago, you know, ten years from now, we're going to look back at today and think, you know, that was zero. >> Yeah, way back in the day there were just PCs and servers and there was traditional IT with rate, today with autonomous cars and IOT and AI and machine learning, it's just going to continue, so that exponential growth that you saw, there's no sign of that slowing down, which is good news for us. >> Yeah, good job security for you for sure. >> You bet! >> Alright Brendan, well, again, thanks for taking a few minutes to sit down and congratulations on the great event and the launch of these new products. >> Thank you, thank you. >> He's Brendan Collins, I'm Jeff Frick, you're watching the Cube from the Western Digital Headquarters in San Jose California. Thanks for watching.