52:25 Webinar
The Goldilocks Solution: Reimagine your Cloud Journey with Near Edge Bare Metal as-a Service
We will continue the discussion from the January TechTalks - “Connected Cloud with Azure & Equinix for EDA & HPC” and dive deeper into Bare Metal as-a-Service.
This webinar first aired on April 7, 2022
The first 5 minute(s) of our recorded Webinars are open; however, if you are enjoying them, we’ll ask for a little information to finish watching.
Click to View Transcript
Good morning everyone and thank you for joining us today for our weekly pure storage tech talks webinar. Um today's event is called the Goldilocks solution Reimagine your cloud journey with near edge bare metal as a service. Now let me introduce our speakers here from pure storage. We have Jack Hogan vice president of Technology Strategy joining us from southern California
and we have Andrew Miller Principal technology strategist joining from south Carolina. We also have our solutions architect rob quest on the line to answer any Q and a written Q and a questions throughout. So now I will turn it over to Jack to get started. Thank you Emily. So I, I really appreciate that everyone's here today.
I want to start off by giving a little background of who I am, who Andrew is and he'll he'll give a little bit of background on himself, but I am the Vice President of Technology Strategy for Pure, what is that title? Don't know, it means that I actually focus on what Pure is developing, but the area I focus on is the cloud and I have a lot of experience
here because for 20 years prior to coming to Pure, I was a C T O for a cloud company on the other side. I was actually consuming technology to make my cloud journey. My cloud capabilities work in my business. Um and so I worked with a lot of Pure his customers and partners and really drive the strategy of where pure is solving things for the cloud journey and so I want to introduce
Andrew, have him cover himself and we'll jump into the topic today. Andrew. Great to have you awesome. Likewise Jack, So Andrew Miller Principal technology here here. Uh Jack and I have some similarities in that start on the customer side for about seven years down in the trenches, a little bit kind of admin architect,
engineer trying to keep ahead operationally and then build stuff and then architect stuff. Partner side for about eight years, build out a technical marketing team. That's where I spent a lot of time in Silicon Valley. I'm just getting to know folks out there and I've been up here for about three years here and really enjoying kind of the role as a it's a semi defined role,
working with customers, lays into product management and engineering, getting to do stuff like this, which is fun. And you know, Jack and I have actually uh I think we first met in person 2.5 years ago, but you'd already been watching some of our videos and I've seen stuff you were doing. So, you know, it's fun. Think about this topic though.
Um you and I, we got to have a little bit of fun with the title. I don't, I think this may be the first time that I've seen Goldilocks in a webinar title, but the idea was, and I did some deep deep research via Wikipedia, but when you actually look at Goldilocks, there's even, I'm pretty sure everyone who sees this as like we think of the idea of just right,
there was actually some previous versions of the story by the way? It wasn't a little girl girl. Goldilocks was actually um, a not so polite old woman who actually broke into a house, you actually see some shades of that and even the story, if you ever told it to your kids and then it turned into Goldilocks,
then it turned into papa bear, mama bear, baby bear. But the idea of, and that's a little bit just a side. We're gonna have fun with us today, not just to take all the time, but the key piece here is that often when we're looking for solutions, we find things that are all the way on this side, All the way on this side,
we need something that's just right in the middle, especially when we're thinking about cloud, public, cloud, private cloud, hybrid, cloud, there's all these different ways and what we're seeing and this is frankly, I think what you've dedicated a good bit of the last couple of years of your life to Jack is, there's almost this kind of in between bridge Goldilocks option before we dive in though, we wanted to make sure to involve you all
with a poll question or two or three actually and actually wanted to get your input. So, uh, any thoughts on these questions, Jack. So how would you answer them if you were sitting in the audience, you know? Well, so first of all I talked to a lot of our customers so you know, I I myself have been there but you know, there is in the concept of the cloud,
everyone wants out of their data center. They want out of the actual physical place. So I would probably say that first question. We're going to see probably a lot of yeses, but recent analysts have said that that a lot of workloads are gonna stay in a data center, but what is the data center? That's what's changing here? And some of the things I think we're gonna cover off today,
it doesn't always involve a public cloud. Doesn't always involve a hyper scaler. Doesn't always involve actually even getting out of managing the physical hardware, but there's different ways that we can solve things and then we'll get into that a little bit later. But if we look at the the poll results, if we can probably kick over to that now.
Alright, let's see sharing those results, you know, this is an interesting thing, you know, and I think what do you see them in here? I mean you talked to a lot of our customers to what you see in these results. I love, I love where you started sometimes. It's this always takes me back to like even high school debate, like definitional questions like what is the data center,
what does that really mean? We we've all like most of us like done some of the journey, probably physical data centers to Volos, but that's been going on for a while, but it's still a data center, but it's at a data center in other ways. So that I mean we're going to play with the definitions here as well as that, primarily migrating to public cloud is number one on question.
The second question there, but it's not all that far ahead of not migrating or even going to service providers and then that last one um, ironically we're just getting started, which is fascinating because if you only read industry news, it's not getting started. We're way down that path. But I think actually that we're just getting started actually matches what we hear when we
talk to customers. Yeah, the that those results and I think we'll get into a couple of examples of that a little later of exactly the results here. So it's good to see. I think we are, we're talking to the right people because I think we've got some answers here, but let's let's go ahead and get into the agenda here and why don't why don't you take us through kind of the this agenda and where we're
gonna roll today. Happy to. So we um, where it says cloud, it's a journey at the top. I didn't, we didn't just edit the slides like red red based on the poll results actually already had that in there because that is what we see. But we want to start off was actually not by talking about pure stuff,
we're gonna get there too. And our partnerships, we want to start with what we're hearing from you as we're having customer discussions and conversations etcetera. And then we're gonna actually roll into a little bit of as we explore this. There's some real things about perception versus reality, how far things really are along, how hard some of this is.
And some of the challenges there are not just technology. Sometimes it's easy to focus on those, but it's also a human thing. If you've ever seen, you know, the classic Venn diagram of people process technology and the three kind of overlap, right? We need technology. We need human involvement and comprehension and processes and opera opera,
opera operational izing my first love for the day, but hey, hopefully it's the last one and then last but not least is as we've looked at this as pure, we've been thinking about what is a Goldilocks solution, something that can sit in the middle that will satisfy a lot of the needs of yesterday in some ways or the stuff that we still need to have their still relevant just because we've done these things for a while
doesn't mean they're old. Just like it still matters to the business. And how about how do we also move forward last but not least like it says Q. And A throughout please don't hesitate to put questions into Q and A. If you want it to be truly answered as a Q and a. Make sure to click the Q and a button at the bottom of the window.
We'll be watching the chat a little bit to live, trying to keep it actually as a, you know, a little bit interactive rob has joined us and he's going to be typing his heart out for Q and A questions and we do also have a spot at the very end to be able to help with that. We will try and do some Q and a questions live, even if he's answering them, you know, well, kind of,
well, kind of pontificate and rough a little bit. But I think into the first section, Jack, well let's get into real life stories here. And let's let's talk about like, you know, we're out in the field, we talk to our customers, we worked with them every day and I love to ask some interesting questions of our customers that are similar to the polls that we have going today.
That the first question is, are you on a cloud journey? And I often hear yes, we're on a journey. But then I asked, where are you in that journey and how far along the path are you and I've worked with, I can tell you some of the largest Fortune 500 brands. And even then when I asked there, see IOS I say, where are you in your cloud journey?
And I get something very similar to what we saw in those, those poll results. I get a lot of them saying Well we're 40% of the way there or were 50% of the way, there were 60% of the way there. And I said that's fantastic because it means you're driving new capabilities and flexibility for your business. You're able to actually respond to business
needs really fast. But then I asked them the uncomfortable questions, What are you gonna do about the 60% or 50% or 40% that you're not already there? What's the challenge there? And that's where I get a lot of the, we're not exactly sure. Um Andrew, you've got a couple of stories here too of maybe some more specific examples of
what does that actually look like with some of our customers. So, so there's one here that I've got a heavily anonymized and this is more about, it was a little painful at the time. Looking back. It's pretty funny and I'm not here to make make it awkward for anybody. So I was having me with a large um large company that you would know the name of it.
I mean you already know Jack, but if you're listening you would know the name if I said it kind of thing. But I was actually called in to actually help do a deep dive on cloud block store, which is another cloud, one of the cloud offerings we have in our portfolio. This is like 22 plus years ago.
And we talked with the kind of director level down managers, directors, etcetera. And they said, you know, hey, this actually we think that what you're going to talk about here really fits what we need and we want to bring our VP in to be able to go and have you walk through this and so it's a deep dive but it's up to york charts so you shouldn't get crazy technical unless you're
asked, okay, you know, respecting your audience. So I go through this and get good feedback and there's some discussion at the very end that the vice president says, Hey, this is really cool, I appreciate your time. I'm feeling good. And then he says, yeah, but actually we're about 70% migrated to the cloud already.
And what I'm thinking is in the call that we had previously with his team. They said they were about 20% migrated to the cloud. So it's this extremely awkward moment where I'm sitting there saying like, do I say anything Like, I don't know that it's my place to tell the VP that his team told me that your 20% migrated and he's thinking there's 70% migrated and I,
it's not the only time we've seen that that was the kind of most stark example, especially where at the moment I'm like what do I say? But we've seen this regularly with customers where sometimes it's disconnected inside organizations but it's also it's a journey and it's actually hard sometimes I think a little bit about of the almost a little bit of peeling an onion,
right? You know, it's the the the happy cloud or application or data center onion, you don't know how many layers there are and each layer that you peel back, you kind of cry a little bit more. It's like there's another complexity layer there. I think that even move moves into a little bit of kind of the the perception versus reality
that we see there we see there Jack. I don't want to cut you off though on any other stories you want to toss in. No listen, I think we hear these stories quite a bit and when you start to drill in and then you start to have had that same conversation with the number of C. IOS of that. That where are we really in the cloud journey?
Because I think just as we saw in some of the poll questions there is this concept of where our what is a cloud? Like what are we trying to get out of a cloud? What what is it? Well, if you take the traditional definition in many people's minds, it's the form of a public hyper scaler. The cloud is Aws or Azure or G.
C. P. Or or Alibaba or all of those capable hyper scholars. But when you then look at the reality of that just last year, when Andy Jassy was still ceo of AWS, he actually came out and publicly said That only 5% of the world's enterprise workloads have moved to a public hyper scaler 5%. That means 95% haven't really moved to.
What many people think is the definition of the cloud. So it really is, it's this perception of what is a cloud, where do you, where do you need to land and what are you trying to get out of it? But you know, and you, you, you have a lot of kind of real conversations with our, with our customers and, and uh, and you know, see what is,
what are some things you see here. So I'm gonna take that one story previously and pull it forward about two years because when you and I were, let's be real, we prepare for this. It's prepared but flexible. So we, I was sharing that story with you and said, I actually talked with that same company and two years later they were saying it's a 3 to 5 year journey.
So fortunately some of the internal communications stuff got, got hired out a little bit. I don't know exactly how it happened, but that, that perception reality of we're going to get going and then it turns into a journey and then Andy Jassy, when you, when you're talking about Andy Jassy, so at the point where he said that he was the head of AWS since then he's going to become
more important. Jeff Bezos has wandered off to do rockets and all kinds of other cool stuff like Bezos gets to do whatever he jolly well wants to at this point in his life, right? But Andy Jassy is over all of amazon. And so the new Ceo for AWS is Adam, so lipsky And he expanded on that recently to say you know that 15% of all IT spending is in cloud, Which is fascinating to me just because that
means that there's 85% that's not in the cloud. When when Adam says this, he doesn't actually say it as a, oh this is a bad thing, like we haven't got enough adoptions. Like look at how much opportunity there still is. But then we can also flip it around and say if we've been talking about this and thinking
about it as practitioners as an industry as uh industry trade journals for however long, years. there's gotta be something going on here more or else it would have been all done. We've been all be off to the next big race and the next new big theme in buzzword in the next new hype cycle, you know kind of thing but you think about even I'm going to pull in a little bit of Gartner
hype cycle kind of stuff, right? You know what actually, I think some were wandering past the trough of disillusionment here. I'm not gonna explain the whole cycle, you can go google it while we're talking right now, we've wandered past some of that and we're starting to get to the point where a lot of the early adoptive workloads are there, we've gotten past some of the oh,
it's hard problems and now we're starting to move into what can we really do and that's where then it becomes and part of that is not just a technology thing, but it's also a human thing because these are different skill sets, these are different procedures. A little bit of, I think I'm and I'm sure maybe it's you, it's been me in a previous time where you see a
trend coming and you want to almost kind of be Gandalf style standing in front, like thou shalt not pass, it's not gonna happen. And then you're like, uh well, I need to figure out how to deal with it. So it's a technology side, but it's also a human piece to, right, Jack. Yeah. So what's what's really interesting when you,
when you look at some of the things that we we've already covered and this perception versus reality, in the public cloud, in, in in your own data center, creating a private cloud in somewhere in between with managed service providers. What is often behind this is the fact that the skill sets that are required to run traditional workloads vary quite a bit from those that run cloud Native deV abs,
you know, ci cd pipelines and all the things that we want to get to in that cloud native. And there's a big gap between there and so it's not just about moving from one location to another. Like I'm out of my data center now, I'm in, you know, public diverse Scaler and now I've solved all my problems like no, I've got a people problem and so that people problem also plays out in the skills that are
in the public cloud are in high demand, you know what, what's the latest number Andrew like how is how is amazon bringing, bringing their folks on now and like who are they, what are they paying them? So there's even some fun stuff. I've got some friends at AWS actually one of them was the first one who sent me, like there's an article recently about AWS doubling
their max pay. So, so by the way, if you're listening in and you're looking for career advice, hopefully already feeling there's there's a trend here, there's probably trends around security and ransomware and other stuff, but this is a good space to be in but publicly Or at least it was in public articles that aws doubled their max base pay the $350,000, some of that is because I think their,
their stock price is still doing well, but it's not the rocket ship that it was in the past, they used to do a lot. But if you talk with folks in the industry, amazon used to a lot of stock compensation, they're competing in other ways. So there's a little bit of a career coaching note. It's a great time to be in this industry if you're looking for what's next and you're
thinking about it and basically where you are in your career. But also if you're a hiring manager, if you're a manager and you're trying to figure out how do I find people, either you have capable smart people in your organization that learn stuff, they're going to have other opportunities to go other places at very different pay scales or it's going to be hard to hire.
Which is this interesting. It's not just even like a people thing of like, oh people don't want to learn stuff. The people that do learn these things, they're majorly in demand and it can be hard to keep them around the kind of thing. I know that when you were um see I oh I think you're probably looking regularly almost attrition rates and how do we keep people happy and all that kind of stuff and say the talent
acquisition is one of the one of the biggest expenses that that I had when I was on that side. So it's, it's really uh it's a real challenge. But you know, the other thing is having been kind of out of sea level position and, and understanding like when the board says we're going to the cloud, what's driving that? Well, it's really a bit of fomo,
right? Like this is like everyone else is going to the cloud, their businesses are reacting really quick and oh, you know, x, y Z. Friend of mine at the club said that, you know, we can get to the cloud in in 60 days and the reality is then hit the teams that they have and the expenses and all of those things. But you know,
there's there's another piece of this which is that shiny picture that, you know, everyone likes to paint that. I've got it under control. Like your story about that customer earlier, like there's probably a little bit of fomo there, right? Andrew there is. So I think um this is actually one thing I was at uh actually speaker.
It was an oracle open world a couple of years ago, I was there to speak and I was walking down the hallway, I don't have time to put my head into one session and I was at a startup at the time. So I actually put my head in and it was like um finance and evolution for startups kind of thing. And it was like and the thing that stuck with me out of that was the average length of a C.
I. O and of a CFO and that we're all human. Right. We're trying to figure out how do we make ourselves look good? Hopefully in ways that align to the organization, you know, and I think that in like just selfish, bad ways. But so if you have the average lifespan for a
CFO and a C I O is usually in the 3 to 4 year range. So they're looking for projects that will make them look good. And if they see the equivalent of, you know, a picture on facebook that looks like there's the cool, cool cloud party out here and they know they're two or three years in and they need something on their resume. There's definitely fear of missing out fomo for
board, board board directors and see IOS I think like, oh, these are these super lofty people, They're human too. And then it even gets to like, um, we, I mean if we don't laugh, we cry sometimes. So if we laugh in any way in in tech, it's probably with Gilbert, right?
So I'll read this to you just in case you're dialed in and you can't actually see it. So, you know, there's the, there's the pointy haired boss, ph b in dover terms, I need to know I'm moving her after the cloud doesn't automatically solve all our problems. I won't ask you to do a double impersonation Jack, I feel like that would be unfair. So you wouldn't let me re architect to be a
cloud native, just put it in containers. I mean we can't solve a problem by saying techie things, kubernetes, we need to do that in a good W W E announcer voice. I don't know. I'll try it again later. But but when we see this, like it's the, it's almost, we're going to kind of spread peanut butter style, you know, we've got lots of holes in the infrastructure things to do.
So there's a new buzzword, let's just spread that all across and hope that enough works out that it kind of looks good. Right? You know, mhm. Back to you. Well, so before we uh, we get into actually how we're solving things here in pure, we've got one final poll question.
You want to lead us into this poll question Andrew. Yeah, definitely, definitely happy to. So if you want to kind of toss that up. So this is now we decided to have a little bit of fun with it. We, we might have talked about this statistic that you may or may not believe Andy Jassy you're adam slips key because hey, they think about Aws and maybe Gcpd and Azure and
Microsoft and oracle have different opinions on this. So you may be like, why are you putting on this poll question? You just talked about it. Yeah, we figured to have some fun with it and see, see if you don't agree with it too. You know, that's cool too. So I see stuff flowing in here,
Jack. We're actually getting higher response on this one than the last one. You know what's, what's, what's interesting about the results here is, and we shared, we shared the answer not five minutes ago, but what's interesting is that I'm seeing is the results are coming in and Emily, why don't we go ahead and share the results?
Because I think that audience will appreciate this, the general perception from the general population. Every time I ask just a general audience, I step up in front of folks and they say, what percentage of the world has moved to a cloud? And this is a really common answer. A lot of people think 50% 30% but even those cloud leaders themselves are saying, no, the public cloud has only brought in 5% 5-15,
depending on the way you look at it. You want to look at enterprise workloads, you want to look at things like sass and other things. So the reality is you know, there is a mind gap that, that, that exists here. Um, and so with that, um, maybe I'll just take us straight into kind of this piece,
but you wanna talk a little bit more about bears. You've already given us one version of the other kind of guiding principles before you kind of jump into the punch line here. I think we'll stick with the fully wholesome one where there's a cute little girl goldilocks that just happens to wander in uh, to an abandoned house and there's absolutely no
breaking and entering because we get to the idea of that, you know, there's too hot, there's too cold and there's just right and, and, and continuing to key off the pole for just a second. Um, the, you know, when we look at the public cloud providers that are actually saying this, that they don't see this as a negative, they don't see that,
you know, they're in the 5 to 15% range is a problem. They have huge businesses, they're driving immense revenue of amazing profit margins and they see it as a huge amount of opportunity that's still out there. But on the flip side there's gotta be some reasons that we are only this far along after how many years of beating the beating the drum on this topic as an industry.
What pure is done with this is we've got a pretty big cloud portfolio, But I think even you're gonna kind of unpack some of the reasons that people haven't been able to get as far as that. 30 or 50% jack Well. And that's exactly where we want to kind of look at the, the past, present and future challenges that every I I.
T. Organization, every company, every ci o faces. I break them down. Is that the past is really your, your legacy applications, those, those applications that have to be running on bare metal servers, traditional 33 tier architectures that that need large databases that have huge amounts of compute processing just in a single vertical solution,
even Vm ware and and VM solutions like they don't lift and shift to the public cloud and I think this is that past challenge but there's also the current challenges and these are all challenges by the way that exists at the same time. Um you know I always like to think that you know the the core mantra of I. T. Is if it works broken why fix it well because
you have challenges like the current day challenges where your high volume big data, your board asking you or your executives asking for more insight from that data. So things like ai and ml large scale scale scale data warehouses that have to run on infrastructure somewhere. These are present challenges but of course everybody wants to get to that future state that cloud native everything is as code.
You can just spin up the next business solution by deploying things like kubernetes or cloud native services. But the reality is that our our our customers and and and everyone we talked to has to deal with all three of these at the same time. And on top of that they have a lot of the challenges that we've kind of touched on in the opening which is why would they build and design and manage things on their own and when
they make the jump to the public cloud or they try, they run into cost overruns because their applications, those legacy applications, they don't report well or, or they need to be completely re architected, which means the business has to stop re architect them and rebuild them. And that's that legacy technology challenge that drags companies back into the past that,
that they've got to run their data center. I think that poll that we started at the beginning is telling a lot of this like, hey, I've got to hang on to my data center because I've got some key apps and then we've really talked a lot about already some of the skills gaps that exist of retraining and moving people into cloud native constructs is not an easy task. And so what you know what we're doing with pure
Yeah, you have a point on that and toss one thing and so the, the application re factoring one to me is a fascinating one. Just because if you have something that's running and this is where, so when I started out actually as a customer, um, and I was just out of college. Okay. And we had a, we had a mainframe environment
still ran a lot of business logic and when I joined the I. T. Organization there, that the main, the timeline to be off the mainframe was five years When I left seven years later, The timeline to be off the mainframe was five years because there's been so much work done but there was still that much more to and this isn't saying that we're going to go down
mainframe paths here. I'm just illustrating from, you know, 20 years ago me because hey, I've lost here and I'm, I'm older than I was then. But the, but the reality is that once we start to get into redoing applications that have business logic versus new projects, it's really hard to justify the opportunity cost when there's actually no net change.
What you mentioned is about running it broken versus fixing it or improving it. And that becomes a real challenge organizationally to justify I think though, rob but we're at the point though where we haven't used enough acronyms. Um, I would really like it if you had a brand new shiny acronym. Not any of any of those like cheap ones that get off the street,
but like a $10 analyst validated acronym. Do you, do you have any of those to toss into the mix here for us? It is an analyst term and I do of course it is mas or bare metal as a service as it's described and actually something now that we're having active analyst conversations around this space that sits in between I as infrastructure as a service,
which most people would think of hyper scholars or even platform as a service, which kind of higher up the stack but bare metal as a service. Is that interim space between the data center where you're running things yourself and the this infrastructure as code that we all want to get too. It's actually capable and available today to be able to run those same things that you have in
your data center um, and be able to deploy them in a cloud like easy to consume as a service fashion. That really eliminates a lot of those risks, those little risks of having to re factor those risks of having to retrain so that the next time you come up with an idea you can immediately scale your infrastructure because that's really what the public cloud is providing is that infrastructure underpinning
your, your business outcome and that ability to do that in a simple and easy way with existing staff and existing capabilities And so Andrew this is the new thing, the new acronym bare metal is a service beam as um, it's here. It's actually coming at everyone pretty quick here. But any, any thoughts on adding adding new new acronyms to the world,
I feel like it's, it's a good one and this isn't made up by the way if you're listening. But this is actually, this is a new analyst category looking at uh, it's one that fortunately is more pronounceable and some of the other acronyms out there. So so Bemis rolls off the tongue decently well, I think that what it is really about the rob jack you're not answering questions, answering questions put the questions in the Q.
And A. By the way that was literally what happened rob types that was like. So what this is really about is what are we solving for it? And is this a sometimes if you've been doing this for a while we can all get a little cynical we can choose to be a glass half full. Glass half empty. I think we've all seen sometimes solutions that
are looking for a problem. This is not that though this is the last couple of years of us seeing this in between zone that wasn't being fulfilled. We had a lot of experience and some you know opinionated design on it. So what can we keep going deeper into what we're solving with this?
Absolutely. But before we kind of exactly understand what we're solving you've got to understand everything you're doing in I. T. Is you're looking for an outcome right? So as an I. T. Leader as an I. T. Infrastructure provider. As a as a as someone who works in the I.
T. Organization, even those developers that sit on top of that that write the code everyone is looking to deliver an outcome with I. T. And and that is got to take into consideration the entire stack and when we say the stack there's different layers that start at the bottom of where you actually are physically the data center of the power all the way up into
the data itself. And within that though you can you can stratify it. You've got a hardware layer and you've got a software layer, you've got an end user layer. And so what bare metal as a service is is actually that hardware layer and delivered in a fully hosted as a service manner. Now what makes bare metal as a service different than infrastructures of service?
Well, quite a bit. It's giving you the same level of control, that same level of physical design with servers that are dedicated to you. They're not just a fractionalized VM or a component of a larger mechanism that you don't understand the underlying hardware and you may need to oversubscribed CPUS. You may need to bring licenses, they're socket based.
Uh, you want that single tenancy control because maybe you have the the need of regulatory compliance issues or data sovereignty issues. Um, but you get all of that in your own control. And so it's just like your own data center where you have root level administrative access. These devices are dedicated to you. Even the providers that we work with and and even pure doesn't have the access to the
information on those systems different than the than what sits in the public cloud, but it's all still available available through that cloud like ease and speed can be fully orchestrated and automated through a P. I. S. Or or that capability to deploy applications on top of it and that's where it really speeds up those development cycles. That ability to leverage things like court works or leverage things like VM ware,
things that you're familiar with or just hey, you wanna raw image of, of Linux or you want to do a pixie boot, you have that level of control just like you do in your data center today, but you no longer have to manage the infrastructure. It's fully managed from a physical basis. Um, all of that is delivered and the consumption of this is all based on the same principles that the public cloud is really
driving, which is aligning the usage to the actual costs. So you only pay for the compute cycles that you, you use and you only pay for the storage as you're writing it pure uses the art technology too, deliver an amazing experience on that side. And so it's pretty, it's pretty unique and I think that's why be masses this category enterprise grade solutions with all this level of control.
I think here we, we've, we've, we've used Goldilocks and we'll wander back around that one. But I also think like I'm hearing you saying that you can have your cake and eat it too a little bit if we could, if we could do that one too because everyone loves cake or at least I do you no good good layer cake that the other piece that I wanted to kind of emphasize here and you already you
already called the saddle. But of course while it doesn't quite show on the slide there, you know, there's a spot where the storage is in in an in an orange box. Right? So there's pure components here. But when you look at what pure provides and probably most of you already pure customers or you're looking at pure, there's a next generation storage capabilities,
whether its snapshots or replication or protections against ransomware. We get into safe mode, there's this whole richness of data services that often when you have environments that are more cloudy, you get the lowest common denominator, a lot of the underlying capabilities, the infrastructure has to be taken away if either for role based access control or multi tenant reasons or there's usually good reasons to do
it. It's not just providers are holding out on you, they have to but this approach means that some of the capabilities that are within that hardware stack can be natively exposed to you. So for instance, um I I talk for better or worse, a lot about ransomware and how anatomy of an attack and then defense in depth and how pure can help with it etcetera.
You know, that kind of thing. When I then wander into bare metal as a service and how we do it. I don't have to put an asterisk on any of the stuff that I'm talking about there it all still applies and that's kind of one microcosm example if you will. But I wanted to highlight that because it's very unique compared to some of the other as a
service offerings where you've got to go and ask like can I can I get a chart to show me all the capabilities that don't apply but I might not know if you didn't tell me ahead of time, that kind of thing. I almost want to I want to push you back and hopefully this won't um I know you're enjoying being a pure but you're a C. I. O. For a while and I'm pretty sure from a C.
I. O standpoint there was a concept around time to value, I don't know if it's like your bonuses were based on it but I know you cared about it a lot, you might kind of playing that out as it relates to bare metal as a service. Yeah. Great setup on that injured thanks. And and I think the reality is every person within I.
T. Is actually living in this time to value space. They just may not realize it. They really may not realize that the whole purpose of I. T. Infrastructure is to serve the business and the faster you can serve the business, the better the outcome. And and that's really the concept of looking at this from a time to value point of view,
the faster something can get deployed, the more money and more opportunity that you have to be able to react to. And there's many times that companies are so far away from this because when they lay down everything that needs to be deployed and you know, I think about the new business ideas, the new acquisitions, we've had that I had in my former roles, I think about new opportunities in the market
that if you can react quickly, you can make a lot of money and it's a huge amount of opportunity. But you also have to think about that stack, that stack that exists that, you know, here on the left hand side of the screen, you've got to think about this as the c I O at least if not the VP of infrastructure all the way down, all of these things have to exist to be able to really achieve that opportunity cost.
And so when we think about it and we kind of lay the individual components that you have to deploy on a time horizon, the most the most challenging things because they're not software, their physical is the infrastructure layer. And I think about, you know, times that I had to deploy data centers, I had to get the space, I had to ultimately contract with co location providers or build
out the heating and cooling within within my own data centers. These things take time, they can take, you know, months quarters and then many times like, you know, a year, year and a half to ultimately make those decisions because they're permanent, that's where the infrastructure sits and then you've got the circuits and the network connectivity so the world can actually access your information or you can access it yourself
and then you've got the individual components, the storage, the compute um and all of them the how you bring it together, what automation, what people and when you lay these things down on traditional time to value kurds, this has been the knock of the actual in data center world, this has been the challenge and the challenge is how how can I solve these things and move
these up? Well that's what we've done in this, in what we're doing with the bare metals of service that are powered by pure, we've worked with top providers like Equinox. Equinox is the world's largest global data center provider. They have the physical space, they also have on demand compute capabilities and they've solved a lot of the big problems of
procurement of that time and we've worked with them to not just automate the infrastructure itself and with things like answerable and terraform, we've solved all of those things the ability to instantaneously spin up compute resources, that ability to rapidly deploy new on demand storage and so it really drives you really far up that curve to really the same place in that curve where your eye as providers are, that's at the same level,
but you have all that control, none of the risk. Your people still know how to operate their environments. And so it's actually a really great solution all cloud like the and the answer is that it checks that box uh in what we do. But this time the value is really what businesses are looking for in their infrastructure and why we're so excited about this.
I think I'm gonna take them looking deeply in your eyes. I think we're doing, we're not even like talking over each other. Look at this, it's amazing. I want to, I want to play with the timepiece there and the time impact and kind of pull that apart a couple of different ways um in a previous life I'll admit to my past sins. I definitely sometimes helped,
you know, 10, 15 years ago, people that had private cloud initiatives like you got VM ware in your data center, let's go check that box together. It's like VM ware is great but it's not necessarily private cloud but sometimes we just had to accomplish it. This is not that this is not just fully hosted, fully managed hardware deployment. This is not put VM are in your data center
because we start to look at where it goes from a, as you mentioned that compressing the down stack pieces and automation but also and I'm also keen to just be real rob, you're out there asking great questions. I don't want to say your last name in case you don't want to be public kind of thing. You know, you've got some great commentary, they're keying off of that.
But as we think about the time impact in it organizations, I know that I lived with the classic, how much time am I keeping the lights on and doing daily maintenance versus new projects or for any new project that I do? How much recurring overhead do I now have baked in? That reduces my new project capability because I did this cool new thing and I know that 5% of me and my team's time is not going to be
maintaining it just just keeping the lights on if you will, but I keep adding more and more lights to my data centers. So, so there's that piece. So it's not just the deployment time, its that continuing operational overhead and then also sometimes the further you get from where you started um the harder it becomes and let me kind of explain that.
So it's like we have ops, we have developers, we say devops like that magically solves the challenge. Not necessarily right because in some ways devops at times as developers fundamentally not wanting to care about ops let's be real, it can be ops as a way to try and handle a new world, but we keep building higher and higher from that layer, that layer cake kind of on the side and use their software hardware.
I'm just drinking where I see this on my screen here. Um, so the more layers that we stack on top, I sometimes think of almost like the force of gravity or weight, like if you if you build a stack of stuff, you put enough heavy things on the top, The bottom layers have to collapse together in some ways or else everything just falls apart. And we've seen this various ways the industry
has tried to deal with this with converged infrastructure, hyper converged infrastructure, trying to take some of these layers and push them together in a way that don't have continuing off road operational overhead, continuing time, you don't have to think about them as discrete things to optimize this goes to some of those same challenges. But in a way that embraces kind of cloud
forward approaches from a consumption and building and all that back to you, rob, Jack and I did it again, it's okay, yeah. You know, and you're right, I think that, you know, time gets really, you know, time really is a big factor that goes just not just beyond how long does it take? Things? I think, you know what we're really showing
here is the collapsing of those times opens up that agility for businesses and really that gets you back to the cloud journey, right? That allows you to reimagine it, this past, present, future challenge, all your legacy apps, all your high volume data driven apps, all your cloud native can now exist in this Goldilocks solution into this mid layer that allows for you to be fully interconnected with all of the
clouds, fully automated and capable of even greater levels of automation into the software layer. Um you know that highly secure capability being actually your own single tenant within a data center that you can identify the assets, those are the ones that are assigned to you and it's really hybrid by nature, that ability to be on a unified platform, unified connectivity,
unified capabilities. Um you know, we're highlighting Equinox here and are pure storage and Equinox metal we are that storage component that's deployed in a cloud console, but what's underlying that is actually the physical device, you're only paying for it as you consume the storage on it. So it really solves all of this because we've integrated it so deeply Um and you know,
it's really uh that kind of perfect scenario to deal with that 85 or 95% of the cloud journey and allows you to rapidly accelerate your plans and get your time back and get that opportunity back in the business. One thing is for for those of you playing playing the at home game, maybe things like on demand consumption that means all the different components have to be.
So if we kind of pull the lens back a little bit in two different ways. One is from a pure standpoint. We've already, we're already doing on demand consumption. You can do everything in a pure portfolio on demand that lines up to Equinix doing on demand stuff. There's also kind of pulling out the lens from, hey we,
this is a multi cloud world. Um, two thoughts there one is whether it's software as a service and we've got our, you know, network from a networking standpoint, we're traversing all over the place to various software as a service offerings. Another one is when I think of when I was at a conference a couple years back and uh, it was actually actually actually at hashtag camp.
So for those, you know, like terraform and console and good stuff there, it's not a free and just being real, we all wander around the industry and the CMO at the time was on, on stage saying. And the quote that stuck with me is that any organization is potentially only one acquisition away from being multi cloud, whether you like it or not.
So we have great stuff that we can talk about with Equinix metal, you can have co location of pure arrays in their native format either to Capex or apex level. Um, you, you can even go and say, hey, you know, I want to have, you know, fleets of stuff, we're doing things with pure fusion. If you want to run legacy applications without re factoring,
you'll see in the kind of the top section there we have pure cloud block store that's a software only offering a flashlight running inside AWS or Azure but that could be very high speed connections like express route and direct, connect back into equinox metal and other equinox resources or you might have pure resources coal load that you want to be able to handle things around. Egress costs.
Like we look at what we do with Azure, Azure has express route direct where instead of paying for egress, you pay at a circuit basis. Well that feels a lot more like it's inside the data center and I'm not tracking every terabyte of data I move, I just, I've got my link kind of thing and of course, last but not least for pure we've got pieces where we heavily leaning into kubernetes and porter works whether that's databases on
demand, that that's what we call pd esport works, data services or a full on kubernetes data services layer on prem in the cloud. So there's a level here we're kind of wanted to pull back the lens a little bit. Thanks for letting me ramble Jack of there's amazing stuff here where we see Equinix metal powered by pure as kind of this if you will goldilocks or you know in between option that is very relevant,
especially if you're getting out of the data center but there's all these other initiatives we have that are wrapping around because this is a huge landscape, 85% of data center spend enterprise spend needs to is going to be going somewhere and there's not um we can get a little bit too if all you have is a hammer, everything looks like a nail. We're trying not to fall victim to that to that
challenge. Yeah. No. And and listen I think that the other element of this Andrew if, if we really look at you know the beauty of this and the full portfolio of where people can help you wherever you are. The other part of this is working with a partner like Equifax allows us to have a global presence um really allows us to have, you know, they are Located in currently the equinox metal capability in 19 global data
centers um and periods right there with you. So that ability to be in new york, Chicago, Dallas silicon valley Frankfurt, London, Hong kong sub Sidney. I mean these are cities that that all allow you to have this level of connectivity so you can build once in one locality and move that exact same kind of automation capability but still with the same level of physical control anywhere else in the world.
So it really is, it is this solution um that is unique and and we're we're really excited to be able to kind of drive this home. Um so and you know as we're getting towards the bottom of the hour here, let's go ahead and you know maybe we, we kind of wrap this up and and get to some Q. And A here. So in case you um let's be real this webinar in
case you have to take a bio break in the middle of your kid, walked into, your dog was barking, kind of tried to wander through, you'll see this in the recording, this is based out of real life and what we're hearing from customers over even the last couple years. Well Jackson, my experience, but what we're hearing from customers because Pierre has already been
making investments in cloud space in general, this is a newer offering. It is capitalizing on that kind of perception versus reality gap because man, there's opportunity there for us and for you too, acknowledging that it's not just a technical challenge. It's also a human side thing from a skill set standpoint and otherwise, and hey, there's an option that can fit right in the middle,
not too hot, not too cold. They're working with their public cloud providers, working well, their data center providers that can kind of sit in the middle, we said Q and A throughout, I appreciate people who have taken us up on that rob has been typing his heart out a little bit here as Jack and I have been talking,
I've got all the names right, you know that like I'm staring at the screen and seeing names pop up, I didn't want to start off with one question. Um, and actually, Jack, this is from a different Robin the chat earlier, but I wanted to kind of put it out, even though rob quist, you've done a great job of answering, but I think it's a good one to kind of pull up here.
There's a question about, you know, is this, is this really just a, and I'm gonna paraphrase here a little bit, um, is this almost just a packaging exercise? Are you just putting putting stuff together? Is it a kind of a fully managed hardware deployment exercise with a fancy lease that's not called a lease, you know, kind of thing, you might kind of unpacking, is that the case or is it something
different? Yeah. And it's, it's an interesting, this goes back to some of the perception things that we talked about earlier. No, it's not. This is actually as simple as it sounds, it's just like consuming the public cloud. You pay as you can get your compute on an as needed basis, literally, you can stand up a physical server,
pay for it on the, on a, by our basis, on the, on the storage side, pure delivers our storage services and you have a minimum term commitment of a year, but it's really allows you that ability to grow, start small scale and grow and so that on demand growth, we, wherever you want to sign up, sign up for years, signed up for two years and anywhere anywhere up to,
you know, five years, you have that ability to really grow with your data and only pay for that as you consume it. So there's no leasing. It's literally you are paying a per gig per month basis on your storage, you're paying a per hour basis for the different nodes of compute just like you would out of an amazon or any of the public cloud. So it's very much different than just packaging
it in a different way. It's truly a consumption based model and it's not and I'm looking at another question and that I'm seeing in the, in the Q and A. That rob is in the middle of answering, but you know, why don't we talk about is this is just pure and colo Andrew is like, is that I mean I'm happy to jump on that, but you want to answer kind of how this is
different than that. I think I'll link the previous question to this one because there was even a great follow up of, you know, doesn't there have to be a premium to cover the physical costs because if it's pure Ndolo, sometimes you have that there is the standpoint of a certain amount of scale, right kind of thing. So Equinox can help from a scale standpoint.
They've got data centers, they cut purchasing power, that's about computing network side at a pier standpoint, if you look at what we've done for a long time and are evergreen capabilities, you can start small and grow non destructively because sometimes there's even the question of like if you're really gonna do this and he is it just pure hello? Well we, because we have the ability from a technical and architecture standpoint to be
able to actually start smaller and grow non destructively and not have the cogs, the, you know, the cost of goods under the covers. So it's not just a Pure and Equinox kind of taking it in the pants from it. So we're just gonna put a bucket of money behind this. There's actually underlying capabilities, especially on the pier side to where we can
start smaller and grow transparently under the covers. And then as that's packaged up, it's not just during a colo um, I've actually been uh impressed you. You're a VP Jack, but you still do good demos of pulling up the equity external environment and within a couple of minutes I've actually seen you go through and like,
hey, I didn't actually have anything ready on the pre pre stage, I just go and deploy what looks like infrastructure as a service level VMS of different types with different storage offerings, that kind of thing. So there of course if it's single tenanted, there has to be some level of reservation under the covers, but there's real full up stack automation with actually a deployment gooey and
a P. I. S. As well as the down stack technical capabilities to help with that, you know, baseline physical cost premium. That is inhibited some other companies from really doing this in a sustainable financial way. Right? And then you get into and you're like, oh, it's not gonna last and then that's a problem too.
So yeah, I and well, and I want to get to this next question. This next one is an interesting one. Can I have flash array and flash blade in the same service? Now, let me step back for a second and explain something. It's a service. Pure delivers Pure as a service. We deliver our flash blade technology as a
service. Today we deliver our flash blade as a service. Today we call it the ufo the unified fast file and object service we call our flash array services are block services are blocked capacity are blocked premium and so it is the same underlying technology. But again, it is a cloud service that gives you the full control of the actual hardware But it is actually consumed the way that you would
consume a traditional cloud storage block or file or even S three. All of those capabilities are fully keep fully available because we're using the underlying pure as a service offerings and capabilities to deliver this to customers. And so, you know, that's really allows you to start in the on premises data center, you can start with that model, that economic model you can consume in this Goldilocks area,
this this fully cloud adjacent hybrid cloud, you know, fully hosted. And so it really gives you all of the benefits of of purest technology. It allows you to consume the entire portfolio. Porter works on top of that. Um, is, you know, another element. So Andrew, I think we're I think we're getting
close, I I didn't want to answer one other question for everybody. What if people want to learn more? Where do we take them next? I think we've got a we might even have a slide for that. Look at that. Just a setup. Right? Yeah. So for anyone who actually wants to learn more
um, and uh, and understand what this what pure is doing here with partners like Equinox and our overall bare metal as a service solutions, it's really simple. Um we've got a Q. R. Code up on the screen, but for those of you that are in listen only mode just go to www dot pure storage dot com slash bare metal. All one word. We make it really simple here at your right
hand. That's like kind of what we're known for www dot dot com slash
Jack Hogan
Vice President, Technology Strategy, Global Alliances, Pure Storage
Andrew Miller
Senior Principal Technologist, Americas
A Goldilocks Solution - Say what?
- Is the public cloud too hot?
- Is on-prem too cold?
- Do you want something just right that fits in-between public cloud & on-prem?
- Are you looking for bare metal levels of control and flexibility with public cloud levels of on-demand provisioning and consumption models?
If you are trying to figure out your cloud journey, you are not alone — and we’re here to help.
Join us as we discuss:
- Cloud Journey - Perception vs. Reality
- FOMO is Real - even at the Board and CIO-level
- Bare Metal as-a-Service, Powered by Pure as a potential Goldilocks “just right” solution between on-prem and public cloud.
We will continue the discussion from the January TechTalks - “Connected Cloud with Azure & Equinix for EDA & HPC” and dive deeper into Bare Metal as-a-Service.
We look forward to having you join Pure Storage for this exciting webinar!
Reduce Risk with Evergreen//One™
Downtime is not an option. Look to Pure to ensure you’re always ready to meet capacity demands.
Meet with an Expert
Let’s talk. Book a 1:1 meeting with one of our experts to discuss your specific needs.
Questions, Comments?
Have a question or comment about Pure products or certifications? We’re here to help.
Schedule a Demo
Schedule a live demo and see for yourself how Pure can help transform your data into powerful outcomes.
Call Sales: 800-976-6494
Media: pr@purestorage.com
Pure Storage, Inc.
2555 Augustine Dr.
Santa Clara, CA 95054
800-379-7873 (general info)