On this week’s episode of the MI&S Datacenter Podcast, hosts Matt, Will, and Paul analyze the week’s top datacenter and datacenter edge news. This week they are talking HPE, Google, AWS, VMware, and more!
Watch the video here:
Listen to the audio here:
4:15 HPE Q4FY24 Earnings
10:49 Willow: A Window On The Multiverse?
18:05 Cloud – The New Silicon Giants
26:19 Observability As A VMware Ripcord?
31:58 It Sees What I See
35:42 Storage Is Cool Again
40:21 Getting To Know Us – Ghost of Christmas Past
HPE Q4FY24 Earnings
https://x.com/WillTownTech/status/1864787679906291962
Willow: A Window On The Multiverse?
https://blog.google/technology/research/google-willow-quantum-chip/
Cloud – The New Silicon Giants
https://moorinsightsstrategy.com/research-notes/some-thoughts-on-aws-reinvent-aws-silicon-and-q-developer/%C3%82
Observability As A VMware Ripcord?
https://www.forbes.com/sites/moorinsights/2024/12/05/the-power-of-deep-observability-in-facilitating-vmware-migrations/
It Sees What I See
https://www.microsoft.com/en-us/microsoft-copilot/blog/2024/12/05/copilot-vision-now-in-preview-a-new-way-to-browse/
Storage Is Cool Again
https://moorinsightsstrategy.com/research-notes/modernizing-your-datacenter-start-with-storage/
Disclaimer: This show is for information and entertainment purposes only. While we will discuss publicly traded companies on this show. The contents of this show should not be taken as investment advice.
Matt Kimball: … the intro video. I think we’re going to end up doing that. I know we’re live but, just for the audience, I think we’re going to end up doing a really cool intro, like a hype video where we all come on and we’re like, “Let’s go.”
Will Townsend: I think we need that, Matt. My Longhorns are in the college football playoffs and they haven’t played well against Georgia, but UT is all about hype videos, so I love that idea. So let’s make that a New Year’s resolution.
Matt Kimball: All right. You got it. You got it. New Year’s resolution. By the way, good morning, good evening, good afternoon, good day to you ladies and gentlemen, boys and girls. We are the mighty three of Moor Insights & Strategy. Mr. Paul Smith-Goodson, wave for us, Paul. There we go. Mr. Will Townsend and Matt Kimball. And we are the Moor Insights & Strategy Data Center Podcast. And we’re here to talk about all things data center, fun, exciting and boring, and not so exciting. So before we get going, just a couple of housekeeping notes. Hit the like and subscribe button, please, whether you like us or not, because we are vain and our egos need to be stroked a little bit.
Will Townsend: We are analysts after all.
Matt Kimball: Exactly. We have big fragile egos. Part two, don’t take anything we say, especially Will, as a basis for making investments or otherwise financial decisions. We are here to give opinions and this is for entertainment purposes only. As you can tell, we are very entertaining. So yeah, really, do not take what we say and call your stockbroker or go on to E-Trade or whatever platforms you use. That would be a bad idea. We are at Episode 34 of the Moor Insights & Strategy Podcast, Data Center Podcast. And I went to look for some exciting and interesting facts about the number 34 because, Paul, I know you like that kind of stuff. It is a boring number. When you type in interesting facts about number 34, nothing comes up. It says here are the first 10 kind of … Anyway, it’s like 34, 68, so on and so forth. Nothing interesting about 34. Not a prime number, not a square root or not a fully square-
Paul Smith-Goodson: Something about every number. I mean, look at it. I mean it’s sequential, it’s odd and even.
Will Townsend: There you go.
Matt Kimball: …today man.
Will Townsend: All right. Paul just knocking it out of the park, man.
Matt Kimball: Some are a little bit more interesting, though, but 34 has nothing going for it. So I’m sorry-
Will Townsend: It could be the number of a famous college or professional football player. I mean I thought … Tony Dorsett was 33 for the Dallas Cowboys.
Matt Kimball: But every number has that, right? So Paul [inaudible 00:03:07] was 34, but every number has that kind of association once you get up into the weird numbers, but what else do you get? Nothing. Anyway, so we’re going to leave it at that. We are going to jump in because we have a lot to talk about and we’re going to squeeze it into a quick amount of time. We are going to be talking about HPE, Google, AWS, VMware, Microsoft, storage in general, and then we’re going to hit on the ghost of Christmas past as a kind of … Yeah, I should have worn a holiday hat but anyway.
Will Townsend: Where’s your headwear? I’m wearing my Mandalorian beanie because it’s below 40 degrees Fahrenheit today. Paul’s looking very sporty with his Nike running cap. He looks like he’s going to go run a marathon at some point.
Matt Kimball: I’m holding on to the last vestiges of hair.
Will Townsend: I have none.
Matt Kimball: By choice. I should have none, but you know what? As bad as I look now, I look worse with a hat. So I’ll just leave it at that. So anyway, we are going to jump right in. And the first one starts with Mr. Will Townsend. We’re going to talk about HPE’s warnings or, warnings, earnings.
Will Townsend: Yes. Well, there’s been a warning and the Juniper acquisition close is pushed out. But for all the pundits out there that are trying to read the tea leaves on that one, the company stated that they didn’t expect for the transaction to close until 2025. And then, there was some momentum late in the year but, long story short, we’ll get to that in a moment. But let’s talk a little bit about earnings. And I know, Matt and Paul, you’ll want to add to this as well. Some interesting highlights. Number one, quarterly revenue rung in at a record for top line and guidance is so important to Wall Street these days. We’ve seen Nvidia just get crushed on that area. But, from a guidance perspective, total revenue, earnings per share and free cash flow for the fiscal year came in all above guidance. So that was super, super positive.
However, my area of focus, one of my areas of focus, networking for the company was down. And this isn’t anything that’s unique to HPE. In general, networking infrastructure is down. But I believe when this deal does close and we know most people expect it to close next year, and I think I’ve talked about this before, combining the engineering capabilities of both Juniper and HPE is potentially very powerful. And then, when you look at the applied research that both companies are doing, certainly HPE Labs, without question, is doing a lot within sustainability. Juniper has been focused as well on their Beyond Labs efforts. So I think that really uniquely positions them.
And I did have, with Patrick Moorhead, our fearless leader, an opportunity to talk to Antonio Neri before earnings were announced. We talked about sustainable architecture and, Matt, if you remember at Discover earlier this year, the labs team was talking about that applied research. And, boy, I tell you what, if HPE can kind of crack the code on that, because that’s a big concern right now with these AI workloads, the energy consumption, I think that could really position them for future success. But what’s your take? Because I know there’s a lot of compute and storage that’s wrapped up into this as well.
Matt Kimball: Yeah, super strong numbers. I’ll go quickly and let Paul jump in with his take. Super strong numbers on the server storage side. Server I think came in at four and three quarters billion, I believe, 31% growth year over year. Fueled largely by AI and, further, fueled largely by NVIDIA, but their margins were up a little bit. So that was positive. And storage, which falls under hybrid cloud, so the hybrid cloud number was up. I think it was 1.5 billion up around 18, 19%, strong growth, which is good because that BU had been down for the last couple of quarters and it’s something we had seen across all of the OEMs.
So super strong numbers. The guide was very strong on both fronts and they had some good announcements they made at Discover around storage as well. So a lot of good momentum that they have. And I think it’s attributed to what you’re saying too. They’re just a company that is … We talk about this all the time with them, right? They are the science project kids and they do really great stuff. And what’s happening now is we’re living in an era where the line between science fiction and reality has become a lot shorter and a lot straighter. So a lot of what they’re doing has a lot more relevance sooner in the market than maybe it did in the past.
Will Townsend: Yeah, and I think what they’re doing with silicon diversity as well is setting them apart from their competitors, their traditional competitors, and Lenovo and Dell as well. But, Paul, what are your thoughts?
Paul Smith-Goodson: I’ll just make kind of a general comment here about looking at large scale acquisitions and whatnot, everything looks really good and everything, but when you’ve got R&D agendas, there’s risk, there’s challenges with integration, cultural alignment, things you’ve got to worry about. It’s just a lot of things that you have to worry about that can go sideways. So it’s just kind of a word of caution when everything looks good. Usually, in my case, when something really looks good, something jumps up in the road and derails it. So HPE needs to deliver on incremental improvements and some really differentiated solutions, I think.
Matt Kimball: Yeah. I do think the sustainability stuff is interesting. I think each one of the OEMs is going to have a really good play around sustainability. Lenovo talks up Neptune quite a bit. Dell is doing a lot around sustainability. HPE is obviously. Yeah, you’re right. More and more they’re kind of and it’s building in. But I do believe, on the sustainability side, what’s making it a lot more interesting to IT now is around cost savings, cost savings, cost savings, and freed up power budgets that were being consumed previously. Right? So finding more ways to squeeze every watt or make best use of every watt. Saving mother earth is great and everybody wants to do that, but the pragmatist in me says IT is driven by-
Will Townsend: Yeah, CapEx and OpEx savings, right? Yeah, absolutely.
Paul Smith-Goodson: There’s some big players in green computing right now.
Matt Kimball: Yeah, really are. Anyway, good chat. Let’s move on to, I don’t even have to ask who this is, A Window On The Multiverse. What the heck does that mean?
Paul Smith-Goodson: It actually relates to Google’s newest 105-qubit semi-conducting quantum chip. There’s some people that think that quantum computing is done basically through the multiverse because, for instance, this one is going to take … The test they calculated is that what would take this chip 200 seconds or five minutes to run, it would take a supercomputer 10 septillion years. That was what they actually ran on. So there’s some people that think that, in order to do those huge calculations, it has to be done in parallel and it’s probably done in some type of parallel universe. So that’s what that comment was. It was just kind of an interesting comment on what really is a big deal in computing.
This new chip is called Willow and it’s 105 qubits and Google’s got a long heritage of really good chips. They started with one called Foxtail in 2017 and had Universal Chrome in 2018 and then Sycamore was really kind of the big jump. That was 2019. I don’t know if you remember or not, but the issue about whether or not Google had actually achieved quantum supremacy or not, and that was with the Sycamore chip. So the Willow has got …. and qubits in it, and that gives it little faster gates and operations and improves the error rates and it’s got improved coherence times. I think the average before was like 20 microseconds, so it moved up to a hundred microseconds. But that’s actually, even though it’s microseconds, it’s still a lot of quantum… superconducting qubits. The big deal is that the logical qubits can function below what we call the critical quantum error threshold and they’re using logical qubits. There’s a lot of research going on in logical qubits, and that’s where they’ve got an underlying base of physical qubits. They can see the errors as they occur and keep the logical qubits error free. So that’s basically it.
But anyway, not going into a lot of detail, it’s really a good chip. It’s done some amazing things and I think there’s a lot more things coming on the horizon. The underpinnings of this really, I think, came from the Sycamore chip and that was created by John Martinez with his Santa Barbara team of students that worked for him. And a lot of things is based on that. I’ve got a call coming up with John on Friday. I want to know what his opinion is of this, because back four years ago when I talked to him about Sycamore, he described this chip perfectly and he said it would be in five years and here it is.
Matt Kimball: Oh wow. That’s crazy. Paul, a couple of questions for you. First is do you see the work that Google does? I know Microsoft has done some work, and I’m sure AWS has done some work, maybe they haven’t, but do you see all of this kind of research and work that they’re doing as accelerating the path for Quantum? Or is it more just their separate research projects that are going on and really don’t?
Paul Smith-Goodson: No, I think there’s a lot. I think a lot of this kind of goes together because Microsoft and Quantinuum, Microsoft and Atom Computing and some of the other companies are working on this logical qubit scheme. Right now, we’re talking about 105 qubits, but the goal for this one, just to give you an idea, is there’s six different milestones that they’re looking at, and they’re going towards milestone three right now, and there’s six of them. And the sixth milestone, that would essentially be a thousand logical qubits and underneath that would be a million qubits. So that’s kind of a long way away yet. But I think we’re getting close to the knee of the curve and that a lot of this stuff can be accelerated over the next five years.
Matt Kimball: So let me ask you one more question. Does this have any impact on … I think of Google, I think of Microsoft and others and there is the internal use of Quantum that both of these companies and many companies will benefit from, right? Obviously. Is there a quantum as a service capability that kind of becomes available through these?
Paul Smith-Goodson: Yes. They’ve all got quantum, I mean.. Dial up , IBM first came out with it six years ago. You can dial up and access a quantum computer.
Matt Kimball: Did you say dial up?
Paul Smith-Goodson: Yeah, on a phone.
Will Townsend: Like AOL, like you’ve got mail?
Paul Smith-Goodson: Yeah. I’m 56K modem.
Matt Kimball: You’re going to need that 56K modem.
Paul Smith-Goodson: Anyway.
Matt Kimball: But anyway, I mean the market is so immature, right? I’m imagining that an IBM does this. It’s really kind of a research play. I’m thinking like commercial kind of applicability-
Paul Smith-Goodson: That’s what they’re calling useful quantum computing. It doesn’t necessarily mean that it’s fully fault tolerant. It means that you can still do some really interesting research with this. And that’s kind of the point we’re at right now. We’re kind of coming over the curve a little bit and we’re starting to look at the useful things that we can do and applications. So we’re getting there.
Matt Kimball: Okay. Interesting. Right on. Hey, let’s jump in because this is a good segue into the next topic, which is Cloud and the kind of rise of the silicon giants, maybe the new silicon giants, if you will. So let me give a little context, and I’m going to go very quick on this because you know how talkative I can get.
Paul Smith-Goodson: Mm-hmm.
Matt Kimball: Shut up. So AWS reinvent was last week, Marvell was this week, Microsoft Ignite was a couple of weeks ago. And they all … Remove Marvell because they’re not a cloud provider, but both AWS and-
Will Townsend: But they do provide silicon to hyperscalers.
Matt Kimball: Yeah, exactly. That’s why I brought them up. They all put or both put a huge focus on silicon, silicon, silicon. With AWS, it was Trainium 2 and Trainium 3, with some talk about Graviton 4. With Microsoft, it was their hardware security module plus Maya going into production and a little bit more on cobalt. And Google has been in the silicon game for probably 15 years now. So it really kind of got me going on the thought of are we looking at a tectonic, if you will, shift in the market where traditional silicon providers, while they are the mass volume, overwhelming majority of what gets shipped into hyperscale data centers today, are cloud providers becoming their own silicon shops, factories, and with the assist of companies like Marvell and Broadcom?
It’s interesting because and I’ll use AWS as the kind of poster child or the use case, if you will, or case study. AWS, they started with Nitro back, I guess, about 10 years ago to do Nitro system, which did some virtualization offload and security and network acceleration. From there, they went into Graviton back in 2018 or 2019 and, in five years, Graviton has gone from, “Let’s see if this thing will fly” into 50% of their internal workloads running on Graviton and nearly half of their public-facing instances running on Graviton. That’s huge, right? And they’re kind of the leading edge of where I see this going. It’s got a huge impact on AMD and Intel. That’s part one. Part two to what you mentioned, Will, about Marvell, they rely on third party kind of custom silicon shops to build very bespoke and very custom platforms connectivity for networking, for security offload, for-
Will Townsend: And also provide the software stack because Marvell doesn’t do that, unlike AMD with Pensando that provides sort of the whole stack with software and silicon, right?
Matt Kimball: Yep. So, as these hyperscalers become, and they are the bulk of the consumption of this technology, it makes you wonder what the path forward looks like. That’s not saying Intel or AMD are going away, or Nvidia for that matter, but boy, it’s really interesting to follow.
Paul Smith-Goodson: If you look at AWS, Trainium 2 and 3, I think they’re basically saying, “We’re not just going to give you a raw compute, we’re going to take the lead on this.” And so, they’re looking at a combination of their AI accelerators and AI driving some of this stuff. And-
Matt Kimball: Oh, go ahead. Sorry Paul, I didn’t mean to interrupt.
Paul Smith-Goodson: Go ahead.
Matt Kimball: I was just going to say Trainium 2, that’s actually…, strangely enough, but when you look at Trainium 2 relative to what they have deployed in the cloud for training today at AWS, it’s the H200 from NVIDIA. They’re claiming 30 to 40% price performance advantage over. And when you look at Trainium 3, which will be … developed, so AWS internal developed from the ground up, they’re saying it’s going to be even orders of magnitude better than Trainium 2. Again, NVIDIA is not going away, but if Trainium 2 or Trainium in general becomes half of what they do from a training perspective and Inferentia, which is their inferencing chip, follows suit, boy, this makes for really interesting dynamics moving forward. And it really makes you kind of wonder where the likes of Intel, AMD, NVIDIA and others, what they evolve into to stay competitive in this space. Interesting, AMD did a custom chip for Azure. That was hosted at Ignite, right? So they’re seeing this approach to got to be a little bit more custom and a little bit more bespoke to run in these clouds.
Paul Smith-Goodson: Customers get a lot of flexibility and, like you said, lower cost and more innovation and a smoother path.
Will Townsend: To me, it seems like it’s an opportunity because all of these big public cloud service providers are building their own compute because that’s a big CapEx savings for them. And then, taking that a step further with silicon, to Paul’s point, there’s got to be some CapEx goodness associated with that. But also I would think, because each of these hyperscalers, they’re good at certain things, and so a lot of customers go multi cloud to get kind of the best combination of features. Each of these hyperscalers are building silicon that really accentuates and accelerates and optimizes what they’re really good at. So would you agree with all of that, Matt, from a compute standpoint?
Matt Kimball: Without a doubt, yeah. There are very little, not little, there are knobs you can tune, but not just knobs you can tune. There are things you can do at the architectural level, how you kind of move data across the compute platform, that can provide significant kind of performance gains for very specific workloads. And, to what you’re saying, AWS and Azure and everybody else are building their own bespoke stacks. So if I can start at the silicon and design silicon that works very specifically with this OS and this virtualizer and this hypervisor I’m putting on top of it, and for the applications that are going to sit on top of there, if I can deliver five, 10, 15% performance gain in that instance, and I can save money on manufacturing and I’m saving even if it’s just a few watts per chip, when you multiply that times the millions of chips deployed, you’re talking about saving almost a data center’s worth of power by doing your own silicon. The ROI is there. Obviously, it’s there because that’s the reason everybody … This is not a “We want to be cool and we want to be smart and do our own silicon.” Everyone, all of these cloud providers, are going hard down the space because there is a lot of money you can save and a lot of value you can add to the experience.
Will Townsend: Yeah. I’m glad you mentioned the OpEx, right, the OpEx piece of that. It’s just not CapEx.
Matt Kimball: No, it’s across the board. Anyway, sorry, we can go on forever. I know I said I was going to talk a lot less. Sorry.
Will Townsend: Come on, Kimball.
Matt Kimball: Sorry. All right. Let’s jump into, now I got the rhythm down, I know this is Will. Let’s jump into the old VMware ripcord.
Will Townsend: I’m going to hit this very quickly. You’ll probably talk more than I will, Matt, on this point, but this is a gratuitous plug for a Forbes piece that I recently published. And what I wrote about was for those VMware customers that are looking for an alternative migration path, how network and data observability solutions can help them on that journey. So in the Forbes article, I talk about that. I highlight Cisco and Gigamon and I also did some research and make some recommendations. You put your eyes on this as well. So this is a gratuitous plug for that Forbes article. It is somewhat controversial but, I mean, what are you hearing these days? Because when Broadcom acquired VMware and call it value extraction, call it whatever you want, there were reports that licensing fees doubled, quadrupled, went up five, six, eight, tenfold, and customers were sort of stuck in a tough spot, right? Because they built their infrastructure, they built all of their strategy around VMware. So what do you think here?
Matt Kimball: Mr. Provocative, huh? This is a good topic. And, by the way, observability, it’s not just a ripcord. It’s a VMware rationalization as well.
Will Townsend: Sure.
Matt Kimball: Right? You want to improve your VMware environment and lower your costs.
Will Townsend: Absolutely. Good point. Yeah, and I point that out too.
Matt Kimball: Yeah, and I think that’s the big thing here. Actually, one of the best things that happened from the Broadcom acquisition of VMware to me is this understanding, this awakening from IT that we get to look at modernization, we get to modernize what we got, whether that is we stay on VMware and upgrade to VCF 9, and we kind of find ways to rationalize our deployment and really kind of lower our cost. Or we go to another platform. Every IT person I speak with is looking at modernization. The further up the stack you go or the further up the segment stack you go, the more tied they are to VMware because of all of the connective tissue and the dependency that have been built over years. The further down, the easier it is to extract but that observability really matters more and more as you go up.
I just want, to what you’re saying on this, Will, AWS at Reinvent announced a service called Q developer for Modernization. And one of the use cases they hit on is you have an on-prem VMware environment and you want to modernize. You want to take that VMware environment, you want to migrate away and you want to migrate up to the cloud in a cloud-native environment. Q developer is like their version of Copilot. So Q developer will go down into your environment, it’ll look at all of your network mappings, look at all your settings, all of your dependencies, co-dependencies, look at your VMware environment. It’ll come back to you and say, “This is what we found and, if you want to upgrade, if you want to take your VMware environment, you want to move up to AWS and at the same time modernize your workloads, click this button.” Click a button, it’ll take your VMware environment-
Will Townsend: Easy button.
Matt Kimball: It’s the easy button. Now, is it a hundred percent? No. But even if it gets you 80% of the way there-
Will Townsend: That’s big.
Matt Kimball: It cuts years off your migration path.
Will Townsend: Yeah.
Matt Kimball: So you’re hitting on something that a lot of companies are starting to look at more and more. I’m sure Azure is going to have a Copilot that does this and GCP will have something, but this is a big deal. And that observability is critical. If you can’t look deeply into your environment and understand where you are and assess where you are today-
Will Townsend: And the interdependencies because it’s so complex, right?
Matt Kimball: Yeah. There’s no way … If you don’t know where you are, you don’t know how to get to where you want to go. So this is critical to understanding where that starting point is. Yeah, so I didn’t mean to hijack your topic.
Will Townsend: No. No, I knew that you’d want to weigh in on that. And thanks for fact-checking my article so I didn’t say anything stupid from a-
Matt Kimball: I didn’t see anything stupid.
Will Townsend: Yeah. When I told Pat I was working on this, Pat’s ears went up because he spent some time with Hock Tan and that team. But hey, listen, this is something that’s on the minds of a lot of folks and I felt it was important to write about it. So again, my gratuitous plug, go hit my social media channels and read up on it and get the number of clicks up for me. I appreciate it.
Matt Kimball: Just the last thing, just to repeat one last thing, and that is this is not just a migrate away from VMware. This is a optimize your VMware environment. There are a lot of paths with this.
Will Townsend: Absolutely.
Matt Kimball: All right. This next one sounds like it’s kind of a horror movie, title of a horror movie.
Paul Smith-Goodson: Yeah. It does, doesn’t it?
Matt Kimball: It Sees What I See.
Will Townsend: Well, Matt, let me just share this with you. So I’ve been traveling the last couple of weeks. I know you have too. And there’ve been a lot of new movies that are on the American Airlines entertainment page that are about AI and it becoming evil. So I was thrilled to see Paul’s sensational headline for this particular subject.
Matt Kimball: Yeah, this has a very, “I’m in your house” kind of feel to it.
Paul Smith-Goodson: Well, this has to do with, obviously, Copilot Vision. So then you can have AI that looks. You got something up on your screen, Microsoft Edge, it’s like a second pair of eyes for you. It’ll scan and analyze and provide insights on whatever web pages that you’ve got up. It helps with planning, shopping, all kinds of things. So it sounds like a simple thing, but I think it’s pretty interesting. I think it’ll have a lot of ramifications. I haven’t tried yet but the early feedback, I guess, is pretty good. It had a bunch of impartial testers, it prioritizes privacy and copyright and gives feedback. I think it’s going to be a success. It’s kind of a natural.
Matt Kimball: Is there a point where it becomes invasive, Paul?
Paul Smith-Goodson: It doesn’t sound like it. It sounds like you’ve got pretty much control over it. And I think you can probably set limits on what you want to do.
Matt Kimball: Oh, that’s cool. That’s very cool.
Paul Smith-Goodson: I think it would be really … I would like to have it.
Matt Kimball: And correct me, I didn’t hear this, so forgive me, but you said this is available now or this is something that’s coming up?
Paul Smith-Goodson: It’s available, I think, for testing right now. It hasn’t been released.
Matt Kimball: Man, it’s crazy to me how much our lives are going to change over the next even just couple of years as AI starts to hit its potential and stride. It’s kind of … Scary is not the right word, but it almost is science fiction-y how different life is going to be for us.
Will Townsend: Yeah, the whole agentic, I guess that’s a new term that’s been coined, right?
Matt Kimball: Yeah, agentic AI and kind of you just sit back and-
Paul Smith-Goodson: Oh, agents are going to be neat. Trump is very pro AI so-
Matt Kimball: It’s beautiful. It’s beautiful.
Will Townsend: It’s great. He’s going to make the AI great again. Let’s not get political. Let’s not go there.
Matt Kimball: I was just doing my dance. I wanted to put on YMCA.
Paul Smith-Goodson: Hold on. I got my animal around here.
Matt Kimball: My animal is at dog camp.
Will Townsend: Executive assistant.
Matt Kimball: You call it your animal. All right. You call it your animal. I call my dog my son. All right. Moving on to the final, Storage Is Cool Again.
Will Townsend: I don’t believe that for a moment.
Matt Kimball: I know you don’t and that’s why you’ll never be a true nerd. So I’ve been working on a number of projects and it struck me how fundamental … This is going to sound like stating the obvious, but how fundamental storage is to everything yet, while we know that kind of mentally or logically, we don’t fully get it. I was having a conversation with Pat a couple of weeks ago, or maybe a week ago, about a project we’re working on, and he said, “Do you really think that storage is the foundation to modernize infrastructure?” He said, “Do you really believe that?” And I’m like, “Yeah.” And he’s like, “You’re crazy.” But I’m not. It’s pretty basic, right? Everything we do today is data-driven.
Paul Smith-Goodson: Storage is like the bedrock of IT.
Matt Kimball: Yes. And you can have the fastest servers with the best accelerators, and you can have the best networking equipment on the market and you can have the best software stacks. But if you’re running on a storage environment that is old and outdated, it is the lowest common … It’s a basic engineering principle, right? The lowest common denominator, the lowest performing part is going to limit what your overall performance looks like. And a lot of companies are investing in a lot of things around AI or data analytics, and they’re overlooking storage.
Will Townsend: Well, they were overlooking networking for a long time.
Paul Smith-Goodson: That’s right. Actually, with AI, I mean data centers are being transformed. They’re not like your grandfather’s data center. You need something that’s flexible, performant, future-proof, because things evolve really fast right now.
Matt Kimball: Yeah. And AI’s going to give way to something else. And there are companies out there, Vast is out there and Vast, they kind of run away from being a storage company. And with good reason. I mean, they’re focusing a lot on data management, data management, data management, but there’s a storage backend to what they do as well. But you’ve got Pure out there that does an amazing job with enterprise kind of storage. You’ve got software defined storage out there, companies like Lightbits that have very innovative software-defined block storage, and they do NVMe over IP. So you get like faster scale and incredible performance for things like transaction processing. The reason I bring this up is if you’re an IT folk person listening to us and you’re looking at how you’re modernizing, and I’m thinking enterprise IT, the data center operators get this, but you enterprise IT folks really think hard about when you’re looking at modernization, kind of where you’re starting from and what those building blocks are. If you’re not starting with what I call the plumbing, which is the networking and the storage, then you’re doing a huge disservice to your organization and you’re never going to really reach the potential you could.
Will Townsend: I agree, yeah. I was just saying, I kind of spoke over Paul earlier, but the focus has been on compute for obvious reasons with GPU and APU and all that. And then networking was sort of the second thought. And now companies like HPE, in my opinion, are taking a real leadership role with respect to focusing on networking. And you could argue Lenovo and Dell and others are doing adequate jobs. But yeah, I mean storage, it’s interesting. That’s where data resides, right? And data is used for training and inference and it’s important.
Matt Kimball: Yep. Long live storage.
Will Townsend: I think it’s cool. I think it’s cool again, Matt. I’ll agree.
Matt Kimball: Thank you. That makes me feel better. Knowing that you think it’s cool is just kind of-
Will Townsend: Thank you.
Matt Kimball: Paul, can I get your affirmation as well, please, validation?
Paul Smith-Goodson: Yeah.
Matt Kimball: All right. Moving on to the last, Getting To Know Us. So we’re doing the Christmas edition. I got a little Paulesque in my title with Ghost of Christmas Past, right? Kind of a nod to A Christmas Carol. So here’s the prompt for you guys. Tell me the most memorable Christmas you can think of in your lives. If you need time to think, I can go first, or if you want me to wait, I can go. Who wants to … Raise your hand if you want to go first.
Will Townsend: I’ll go.
Matt Kimball: Okay.
Will Townsend: Something came to me, Matt, when you mentioned this before we started recording. I’ll make it really, really quick. So I think I was probably eight or nine years old. I have a younger brother, Carl. He’s the smart one in the family. He’s a CPA and he’s a controller for a startup here in Austin. But I remember one Christmas, we were a blended family, so my dad had remarried. And so we had gifts coming from different family members and probably not a lot of coordination. But anyway, long story short, we started opening gifts from other family members. And I opened mine and it’s like a firetruck, right? Super cool. My brother opens his first one, and it’s a hat. It’s a cowboy hat. It’s like, okay, go to my second one. I open it up. It’s something really cool like, I don’t know, maybe Star Wars or something. My brother opens up his second one. It’s another hat. This is no joke. When we were done unwrapping the gifts from kind of the other families, my brother had four hats and was crying. I guess I was eight, he was six. So I remember my mom and dad forcing me to like do a tradesy with him. So I got one of his hats and he got one of my cooler toys. And I don’t know why that’s remained and stuck on the hard drive but yeah, when you mentioned this, it just came to me.
Matt Kimball: I love that.
Will Townsend: Sorry, Carl, sorry that you got gypped when you were six years old.
Matt Kimball: So you took joy in his getting gypped, though.
Will Townsend: I actually because, you know, when you’re eight years old, you take joy in the misery of your siblings, right?
Matt Kimball: Heck, yeah. When you’re the youngest, it’s a survival technique.
Will Townsend: For sure. I was lucky because I was the oldest.
Matt Kimball: Paul, what you got?
Paul Smith-Goodson: I think I was about the same age. I think I was maybe nine or 10. The thing I wanted most in my life, I would have given up just about anything for a chemistry set.
Will Townsend: Of course.
Paul Smith-Goodson: And my parents, I knew they had a … at Christmas time. But anyway, when it came time to open up my present, I was pretty excited but I wasn’t as excited as probably they thought I should be because the package looked like it had been wrapped at the zoo. And then, when I unwrapped it and opened up my little folding metal chemistry set, my dad said, “Geez, it looks like somebody’s been using this. I wonder if they gave us a used set.” They knew I had broken into the vault and…
Matt Kimball: Oh, that is awesome.
Paul Smith-Goodson: I thought I was really being cool.
Will Townsend: Why are we not surprised that Paul wanted a chemistry set?
Matt Kimball: Oh, you didn’t want one? Oh man, that was my favorite gift ever was my chemistry set.
Paul Smith-Goodson: I invisioned mushroom clouds popping up in our backyard. I was really disappointed because I couldn’t make anything really work, though.
Matt Kimball: Paul, did you have a George Washington moment when your dad said that?
Paul Smith-Goodson: No. I thought I was clever enough to avoid that.
Matt Kimball: You should have turned it around. You should have been like, “Did you buy me a used one?”
Paul Smith-Goodson: Yeah.
Will Townsend: Hey, maybe next year I’ll share with you guys how I found out there was no longer a Santa Claus.
Matt Kimball: You can tell it real quickly. This’ll be good.
Will Townsend: Okay. Real quick. So I guess it was a couple of years later, I wanted this Bit Jim action set. He was like the male model version of GI Joe, and I wanted this like ski set, and it’s like this very elaborate snow ski thing. So I open it up, I’m super excited, my dad’s putting it together, and at one point my dad leans over to my mom and goes, “Hey, remind me to send Santa a note that I no longer want to be putting together complicated toys on Christmas Day.”
And it’s like, “Dad, are you freaking kidding me there’s no Santa?”
Matt Kimball: Meltdown. And I’ve been in therapy ever since.
Will Townsend: Yeah, for sure. All right. All right, Matt, you share. Your time to share.
Matt Kimball: All right, so mine’s a weird one but, honestly, Paul, my chemistry set was my favorite gift ever. And man, I loved looking under the microscope at salt crystals and … Anyway, so it was 1983, my senior year of high school. My sisters and I, I’m the youngest of five, and my two older brothers were in the service at the time. My sisters and I went to mass. We went to my grandmother’s to see her after. And we got a phone call from my oldest brother who was in California that said, “Hey, I want to let you know” … this isn’t funny but it kind of is … “the house is on fire and I’ve lost everything.” And I’m like, “Wait a minute, you’re in California, you’re calling us to tell us that the house had burned down?” So we lived in a triple-decker. It’s a Boston style apartment. It’s big and it’s got six families living in it. And they’re old and they’re for poor people. There’s no nice way to put it. So they’re outdated and they’re not well-kept, and so the electrical, the wiring of this triple-decker was from the 1800s, late 1800s, early 1900s. And, apparently, something in the wall caught and the entire house or the tenement burned down and we were homeless for like three weeks.
Paul Smith-Goodson: Oh my God.
Will Townsend: Oh my God.
Matt Kimball: True story. No, it’s kind of funny, right? Because it’s like, “Wait a minute. You’re 3,000 miles away and you’re calling me to tell me?” So what had happened was his friend lived around the corner from us and called my brother who in turn called us. You want to know the irony in all of this? My dad’s a firefighter in that same city and they were maybe two blocks away from us.
Will Townsend: Oh my.
Matt Kimball: Yeah.
Will Townsend: And y’all were at church when it caught on fire.
Matt Kimball: Yeah, my mom wasn’t so my mom didn’t call us. She was at home but she had no way to call us. There were no cell phones back in 1983. And so my dad is there putting out the fire where his kids and ex-wife lived-
Will Townsend: Oh my gosh.
Matt Kimball: Yeah. How’s that for a fun story, huh?
Will Townsend: I think I’m going to go drink some spiked eggnog now.
Matt Kimball: Hey Will, hey Will, that’s when I found out there was no Santa Claus.
Will Townsend: I bet. I mean for me it was Big Jim. For you, it was like your home burning down, Matt. Oh my god.
Matt Kimball: It was the heat miser.
Will Townsend: Oh, that’s terrible. You always one up us on these things.
Matt Kimball: That is a one hundred percent true story. And if any of my siblings watch this, they will be able to … which they don’t.
Paul Smith-Goodson: We should have a “How poor were you?”
Matt Kimball: How poor were we? Oh, you don’t want to know. You want to know how poor we were? In the winter, and this is in Boston in the winter, we would boil water on the stove to get warm water for baths because we didn’t have any oil. So we had no heat, we had no hot water.
Will Townsend: Wow.
Matt Kimball: We were that poor. You want to one up me on that one, Paul?
Paul Smith-Goodson: I can one up you on that.
Matt Kimball: You didn’t have any indoor plumbing?
Paul Smith-Goodson: It was down the hall. We lived, my address when I was young was 1510 High Street. High Street is now Martin Luther King Drive. We were one block away from the black university. My best friend used to be a little black kid that lived down the street. I thought he was rich because he had a house. One day I went to see him, play with him, and he just cracked open the door and he said, “My mom said I can’t play with white trash anymore.”
Matt Kimball: I think my relatives said that to me at some point.
Will Townsend: Oh my gosh. Well, I feel guilty now for living in a middle class household.
Matt Kimball: Yeah. Well, you want to tell us about the four hats you got for Christmas?
Will Townsend: Yeah, and the four cowboy hats I got for Christmas. But my dad was actually born in Brazil on the banks of the Amazon River. It’s a long story. And his father worked for the government, not very wealthy, but my grandfather was determined to send both of my uncles and my dad and my aunt to the States so they could get an education and they could move up in the world.
Matt Kimball: That is very cool. Anyway, all right, we’re going to end on this note because we can go down a really interesting path, but we’re not going to. Listen, before we close out, if you liked what we talked about, please, please, please. Or you like the podcast or just think Paul’s a really cool guy, hit the like button, subscribe. If you don’t, come on back for the next one, it’ll be better, I promise. Although this was pretty darn good. And lastly, again, don’t take anything we say as financial advice or trading advice. If you do, we’re the least of your problems. You’ve got a lot of internal problems you’ve got to deal with. Hope you all have a wonderful Hanukkah. Happy Merry Christmas, Happy Kwanzaa. Whatever your beliefs are and whatever your celebration is, we hope you enjoy it. Happy New Year and we’ll see you guys in 2025.
Will Townsend: See you in ’25.
Matt Kimball: Thanks guys. Have a great day.
Will Townsend: You too, Matt.
Paul Smith-Goodson: See-yah.



