56:07 Webinar
Enterprise Data Cloud: Managing Data, Not Just Storage
In this session, we'll dig into what it means to manage a platform built for both operational performance and AI readiness, and the changes when infrastructure can finally gain visibility into its data, not just storage.
This webinar first aired on August 6, 2026
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
Hello, everybody, and happy Thursday. Welcome to this edition of Tech Talks, and just like we did in the intro, we're gonna be talking about managing data, not just your storage. And to help me with that topic is my good friend, Michael Leworthy. Michael, how are you doing this morning?
Really good, Jason. Mate, how you doing? I'm, you know, I, I'm great. I'm on I'm actually on PTO this week, everybody, but I, I wanted to be here with you and the audience, to hear everything about data s- you know, managing data and everything with Everpure is now doing to make that easier.
I'm a longtime storage nerd. I've been doing this for a few decades. Don't let the face fool you. I'm probably older than you think. So I think it's a good topic, especially as you think about AI or large amounts of data
for databases and workloads and, you know. But obviously AI is the, the elephant in the room as far as, like, how, how do you have all this data? What are you doing this data? How are you using it, simplifying it, all these other things. But we should probably, before we get into all of that, why don't you take a second to tell
everybody who you are? They p- probably know, I know I've seen you around the, the webinar, loop, and I saw you accelerate as well, but maybe some- we got some new folks here, so. Yeah, yeah. So I'm Michael Leworthy. I've been at Pure for, ooh, six years now, I think.
Um- Oh, wow. Congrats. Yeah, yeah. Been, been a while. Long-term Microsoft guy before that. Did a bit of, bit of work at Veeam and stuff, so I got a bit of background in storage and backup and managing data.
Primarily my role here is to help the company with this, their data management strategy. Like, how do we think about 'Cause we have a ton of great products, FlashArray and FlashBlade and- Yeah you know, Purity and Pure1 and Fusion. And they're all really good, but when they all come together it's magic w- with the Everpure platform.
So my team sort of helps, you know, how do we help our customers understand the Everpure platform, and how can you build cool things like EDC and these data privacy things that Charlie talks about. Yeah. And EDC meaning? Enterprise Data Cloud.
There we go. Archi- an architecture to bring all your data together under a unified model. Yeah. Yeah, yeah. We, we, it rolls off our tongue, but maybe we got some folks- Totally rolls off the tongue. Got some folks I did not know you spent some time at Microsoft. We should sync up on that at some point.
Yeah, about, like, 17 years. So just a little bit of time. Wow. Well, we might cross- Just a little we might cross paths. I was there earlier on in my career, but anyways, we can, we can reminisce on that later. Absolutely. Totally.
Well, when we talk about data primacy in the EDC or the Enterprise Data Cloud, like, what, what, what's some of the groundwork or the some of the foundations that we should be starting with that we probably should, you know, layer up so folks get the full vision of as, as our portfolio, right? Like you said, we have great products- Mm-hmm which are great, you know, but when you put
them all together, that's what, you know, to your point, that's where the magic, 'cause I'm doing my hands, the magic happens. So. Yeah, yeah, yeah. I think, you know, I think the big thing is, like, how do you think about architecture and, and the model of architecture and what's changed in the industry?
Because, like, as you know, like, we go way back in IT, and IT hasn't changed a ton over the last 20 years. Yeah, virtualization came along and made things easy. We didn't have big iron anymore. We just had, you know, lots of virtual machines doing the same thing. But essentially we sort of, treated storage and essentially infrastructure in exactly the
same way. Sort of applications owned data, and we built data sets around applications. This is my database application running on block. Here's my file server stuff. Here's my object for, you know, to scale out or, or I'm, I'm building early, you know,
analytic models or machine learning models and things like that. So that hasn't changed for quite some time. And you know, and it just got more and more complex as we just bolt stuff to it. Yeah. You know, the, the whole model had to change because one, just even from a complexity perspective, things became
just too hard to operate. Like, it's, you know, it's hard. I want weekends, I want night times, you know. Mm-hmm. Like, I don't want another system to manage with another set of management, another set of policies, and another set of training and certification and, oh my God, another
forklift upgrade down the road. you know, and sort of two, AI came along and sort of ripped this cover off in a massive way. Yes. You know, and, and I think caused a problem. And I'd love to talk about, I think, the problem that it caused, just because- Yeah if you don't, if you don't mind. Yeah, let's do that.
Let me bring up a q- a slide here. I sort of mentioned before, like, when we think about data and this infrastructure of data, it has been trapped by applications. And that's been a good thing in the past, right? Like, my data- database data should be with my databases.
Yeah. But what happened is, like, I have CRM data, you know, I have, manufacturing data, I've got customer data, I've got database data, I've got all this different data everywhere. And the context of that data is only known to that application. And so essentially sometimes the data, I might have the same set of data in three different
systems, especially for my sales organization. I might have customer-related data, invoicing data, purchasing data, shipping data, all based on Michael Leworthy, and it's all different. Yeah. And, and that's sort of fine if I'm just doing invoicing.
But you know what? Then AI comes along and looks at all that data, and it doesn't know whether that data's real- Right what the most important things are, and it will make a determination based on itself. It will hallucinate. Right. And that's why a, a lot of these AI projects
are, are really struggling right now. You know, I heard this stat the other day that 60% of AI projects are just, a- abandoned because of hallucination. Like, they're just not getting the results they expect out of it. And then, like, most, most organizations, like, over 90% of AI projects are just not getting
the ROI back. They're, they're not getting more out than they're putting in. And that's not a model problem. Like, their models are good. People are building great pipelines and AI models, but they're, they're sort of looking
at it going, "It's gotta be something with the model. It's gotta be something with the model, right?" It's not. It's the data underneath. Yeah. Like- If you've got this fragmented data all underneath, AI just is not your friend in that situation, so- Yeah, it's almost the cliche, and I hate
to say it, the chat will probably cringe, it's garbage in, garbage out, right? Like, I mean, to a point, obviously. Yeah. So 100%. Yeah. And look, that's a good thing. 50% of the data in most organizations, they can't even kn- they can't even find.
It's dark data. Yeah. So that's the whole thing, and then 30% of that data is garbage, right? It's old, unrelated data. AI doesn't care. AI's just gonna go grab it, crunch it, and
then, you know, splurt out a, you know, some sort of result, and then you go, "Oh my God," like, "This is terrible," like, "Why am I doing this?" Right? And but, but my boss is telling me that AI's the future. Yeah. Right. Like, I'm in this I'm stuck. I'm super, super stuck. So like, we keep saying, like, the, the Your
AI models are not the problem. People are building phenomenal AI models. Everyone understands sort of the vectorization of data and how data Like, that sort of They have a good understanding of how AI works, like, from a mathematical tho- theoretical perspective, but the data underneath is not set up for it, and we have to change that.
Like, and so we've sort of had this idea of, how do you get data out of application cycle, so, silos, and put this data in a unified way with context so the data has its own context? Hey, I'm sales data, and I have this context. I ta- this is the person, this is their information, here's their pur- so I have context.
So I don't need the CRM system to tell me that's sales data. That data has its own context. So once I have that, once all data is contextually aware, now I can have that data in a unified environment. Not a physically unified environment like the old data lake stuff.
Mm-hmm. But like, I can have it in a virtualized unified environment. Where it lives physically is, is where it lives, but that data has context, and now I can build applications and agents on top of that. Now AI goes and talks to the data, and the data is contextually aware of itself and tells
AI, "Hey, this is the data about me." So AI doesn't have to do the guessing. The data tells AI what to do. That's honestly the model we need to move to from a data primacy perspective. Okay. That all makes sense, like, theoretically. Like, I get I, I'm buying what you're selling, but, like So I'm assuming
the next step is Exactly. Yeah. So the next step is you've gotta have an architecture to be able to support that. It's a great idea, right? Yeah. Yeah, data that has context, like, this is really good. But again, you can't, like, do this with legacy stuff.
It's not a plug-in. Right You don't, like You know, there's not an infomercial thing you can buy and just apply it, this works overnight. There's gotta be some change here. You gotta change the model, right? And so you need The, the model here is consistency.
I mean, the problem with data primacy, you know, like, when we think about application architecture and application silos, is this is a problem with infrastructure. Infrastructure is just fragmented, right? Because we've built 20, 30 years. It's like, it's like building a house for 20 Building a house, you know, over 20 years with
20 different rooms, and I use 20 different, consultants to do that, right? Like, and architects to do that. I'm gonna have c- 20 complete different rooms, and plumbing's not gonna align, and there's all this problem, right? And that's where we are now, all right? And everybody that's listening to here is just using brute force to solve that problem, and
they've been doing it quite well for a while. But now here, we're in this sort of situation now where if we have a consistent platform, a consistent way to store our data from a unification perspective, doesn't matter, it's block, file, and object, where it is physically, where it is virtually, I interact with it as a single plane of data.
I can then apply global policies across that, so now I think about managing that globally versus system by system by system by volume, which is where I'm usually are right now. So now I, I can interact with that, see that, apply global regulations or apply global, global compliance with that, and now I can then build a universal data intelligence layer across that. So that's where the model we need to speak to,
is like how do we move our architecture into a very consistent way to store our data, govern our data, and then understand our data? Yeah. Yeah, it's You know, you Sorry, I We got some stuff coming in in chat, so I'm like- I'm multi-purposing here, y'all. It's funny you talk about this because, you know, I, I won't speak for everybody, like,
I've been around the storage block a few times, right? And it's You, you talk about things where it's like you mentioned data lakes, right? Like, I remember, you know, mid And, well, I got a question on that, but I remember the mid-2010s, and it's like, because you did You, as an infrastructure person, you had the box or the array for this, and you had- Yes you know, maybe it's a scale-out file or whatever,
and then you had your high-performance box for your database workloads, and that was that box or that array, I should say. Array box is probably a little insulting. But you, you had these very purpose-built systems, but like you said, you know, they never They didn't talk.
You couldn't organize your data that well, and, you know, y- your boss would come to you. I always, I always loved this, it's like, "How much storage are we gonna need next year?" And I'm like Yeah, exactly. I, I, I don't know, a lot? Like Exactly, right? Or We don't know, right?
Like, yeah. Yeah. Yeah. And they're like, "We're gonna give you, like, how much budget do you need in mon- " All of it. Like- Yeah. Like, just tell me everything. So I feel like, you know, this is definitely something that folks have heard and seen and talked about before, but now I feel like what, what you're talking about is, like, we're
actually really doing it. Like, we've, we've, we've It's like the Six Million Dollar Man. I'm aging myself here, but we have the technology to do this. And I think that's, I think that's the key, right? Like, I think, like, if we go back 10 or 15 years ago, the technology wasn't available to
do this because, like, the architecture forced us in this way. We went from big iron, and then we went from virtualization, but sort of the hardware is very, very commodity. It wasn't sort of set up, to do this, and the biggest, the biggest inflection here was c- public cloud. So public cloud taught us a way to abstract
hardware from software at scale, at a mega scale. So we were like, all right, how do you take that mega scale and bring that down into a, a scale that an enterprise can use everywhere, right? So that's the idea behind a thing like Enterprise Data Cloud, is this is a public cloud model. Like, cloud is a model to
me, not a physical place. So like, how do you bring that model and, and make that model available across And there's lots of people that do this. Like Nutanix do this, Azure Stack does this at, from a Microsoft perspective. AppHost does this from an infrastructure perspective. Yeah.
Enterprise Data Cloud is doing this for storage, and for data at our level as well. So it gives the ability for, you know, this organization to abstract this, their data from their hardware and be able to sort of manage that uni- from a unified perspective across everything, right? Yeah. So let's I wanna do something here.
Okay. Producer Laura, I hope you're there. Everybody, we've got Producer Laura, give her a hand. Let's I got a couple questions. So l- what do you feel about doing a quick lightning round on some questions? I love it. I love it. And then we'll get in, get more
under the hood. Love it. Unintended, so. Let's do it. Let's do it. All right. So I'm gonna buy I'm gonna key on something you already said, so I feel like I'm You baited me, so I'm gonna ask. Okay.
And we already talked about the 2010s. How is this different than a data lake? Like, give me the, give me the little bit like, we've been there, we've done that, we've got the T-shirt- Yeah, yeah, yeah in the 2010s. Yeah. At least some of us did.
Yeah. But how, how have we, how have we changed that? Like, what makes this different? All right. Yeah, yeah. You know, da- da- data lake was a great concept, right? Like, yeah, let's put all our data in one area and b- build a taxonomy around it so it c- all these applications, essentially at that stage,
analytics and machine learning can sort of be able to pull that out. It's really hard to move data around in most organizations. Yes. Right? And also, there are regulations and compliance that, that I wanna respect in most organizations, right? Like, I'm going to build physical infrastructure for reasons.
Like, for example, if I'm an FSI organization, I'm holding really sensitive financial information, and there is both digital and physical things that I have to put in place to protect that data. I gotta put scanners on the door to be able to get into the data center, right? Yeah. There are certain regulations from a physical security infrastructure.
So I, I just can't put that financial data in somebody else's pot, right? I wanna control that data physically. There are some data that I'm gonna say, all right, it behooves me to put that in the cloud because it's cheap. Glacial storage in AWS and Azure is actually really economical.
Yeah. Right? I, I don't need access to glacial data, the archive data all that often, so you know, I'm gonna put that in the, in, in, in, in, in Azure, and I'm just gonna have the economic benefits for that. So I'm gonna make decisions based on my needs where physical location of data should live. Regulations, compliance, performance, n- a nearness to my users, all
that type of stuff, right? I think that's where data lake went, sort of like went wrong, right? Where we say, all right, like we respect, put your data physically where you need to be, but now you can build a virtual layer across that so that data looks like it's right there next door to you, right? It doesn't, doesn't care
where it is physically. It now exposes that up in a unified virtual environment through whatever, you know, zones, tiers, whatever taxonomy you wanna build on top of that. Layers. Like what- Right? Exactly. That's completely up to you. Yeah.
But now you have a unified data p- the data plane where I can say, "Oh, well, that's data that's in the San Francisco array 2792. Who cares where it is?" Now that can e- be exposed. You know what? It cuts down copy data management. I can use original data for analytics and AI.
It, it enables sort of all the best of things that what data lake promised, but it respects my need to store data where I need to have it because of geopolitical reasons, data sovereignty reasons, financial reasons, regulation reasons, compliance reasons. Got it. No, that makes, that makes more sense, right? Because it's like, again, got the T-shirt in the 2010s. Mm-hmm. Finally.
So, but let's talk about like scale and scope. Yeah. Like, is this something, you know, that we've, we've got customers, big, small, medium, and you know, those terms mean relative things. But like- Yeah if I've got five arrays or I got 200 arrays, like i- is this fit for everybody or is it kinda more like, "Hey, you've got all these systems." Mm-hmm.
Like, this is a better fit for you? Or kinda talk about if I'm, if I'm, if I'm a pure customer now hearing this, like do I How does this apply to my existing fleet- Yeah. Yeah, yeah no matter the size, I guess- It's a great- is the best way to phrase it. It's a great question, right?
Like, it's absolutely a great question. Like, you know, we, we hear the word enterprise, and does that mean big- Right fragility? Like, you know, like lots of effort, you know? I think one of the great things about this architecture is it's pliable
for any type of organization. So even if you're a, a small customer with two arrays or three arrays, there's a benefit for you to grab those three arrays and put them in a fleet of three. So now you Like now you can ex- like all the capacity of those three arrays is now fully available to you, for you to use, right?
The system will determine where workload placement goes or where volume placement goes. That's, that's gonna be automatic for you. So even if you're a very small organization just with a couple of arrays, the, there is incredible benefit to, for you to apply an Enterprise Data Cloud and like a fleet managed system.
If you've got 200 arrays, 2000 arrays, again, like the economies of scale is really big. Let me give you a really great example. Like right now, a ton of customers are s- are really hurting because of supply chain, right? Right. From a name, price perspective. I'm hurting on my PC game, my gaming PC, so yeah.
Oh my God. Yeah. Like it's everywhere. Actually, my, my kids are crazy right now. They're like, "Hey, birthday's coming up, I need a GPU." I'm like, "Yeah, I can't afford it." You're like, "Let me sell my kidney." I have to take out a HELOC at this stage, right? So, so when we think about, when we sort of think about capacity, like we know in most
organizations storage is reasonably, you know, was reasonably inexpensive. So when I wanted to grow a capacity, it was easy for me to purchase new capacity, right? But now it's really, really hard for me to do that. Expensive and hard. So mo- a lot of organizations are going, "Hey, I need better resource utilization.
I need to move my utilization from 40 or 50% to 80 or 90%. How do I do that?" You know what? That is super hard to do if you've got individual arrays and induvi- individual silos. Like, that is a m- a Excel matrix nightmare to be able to do. What we're saying is- Like the b- the best tracking- Oh tool for systems,
the Excel spreadsheet. 100%, I love it. I do love the Excel sheet. But- Who needs a CMDB when I've got an Excel spreadsheet? Darly, we all, we all do it. We all do it as IT guys. Yeah.
So, but now I've got a system that can come in a- and, and, and just join all those arrays together. Every array knows every other array. They know its capacities, performance, its SLAs. And the system right now expose all that capacity and say, "All right. This is a, a FlashArray that can support, support block file and object.
And here's a FlashBlade that can support, you know, object and file." And I know when I deploy an application now, when I deploy a workload now that requires these amount of SLAs, I can put them on this FlashArray or this FlashArray or, you know, this FlashBlade, and I'm gonna c- c- reconfigure the environment through dynamic workload rebalancing and things like that to ensure that I'm using the utilization, resource
utilization, as well as respecting a- any SLAs. That is huge, just that piece- Yeah is huge, right? Like, just that small piece of enterprise data cloud, which is basically just exposing the capacity, what you have already in the environment. So think about, like, if you're You've got 2,000 arrays.
Oh my God, you could probably just sc- you squeeze so much more capacity out of those arrays, you know, in an, in from an organization perspective. So again, like, this is available for any type of customer, whether you're two arrays where you, where you know it and you're just looking at, a way to actually, a way to actually deploy applications faster without having operational overhead, or if you want m- much
better resource utilization or efficiency or agility. The model is incredibly fungible no matter what size you are. Yeah. I just think, you know, I c- I think about it and I, I know I kinda set up that question 'cause we The, the right answer is it's, you know, even if you have one array- Yeah, absolutely.
If you can au- if you can automate a task or you can make th- like, you It was just gonna, you know, it's gonna make If you're a- an IT manager, it's gonna make your team's life easier. If you're a- Oh, yeah director who cares about the budget, you're gonna get better utilization.
I mean, like if- Like, there's so many- Yeah better things about it, but So the answer is one to a million. One to a million. I mean- I mean, if I could A- anything I could standardize is great. Right. Right? Like, if I can create a blueprint or a
recipe to support an application, that application is gonna be deployed exactly the same way every time. Right. So then drift support becomes so much easier, management. Don't get me to my days of dealing with SQL DBAs, where every SQL Server, no offense if
there's any on here that- Roger like, everything was like, "Oh, I need my drives S and T." And then somebody else is like- No "I use L and V." And I'm like, "Oh my God, you make my life-" Think about if you can remove day two operations out of your manual process and just put that in a blueprint and then it, it's consistent across the board. Even that. Yeah. I'm like, oh my God.
Just I, I would die to have that, I mean, in my life. I'm gonna ask you one more question, and I feel like this is, this for me, this is a big one. Okay. Well, how much is this gonna cost me? Yeah. That's great. Like, I hear all this like, like,
like, you, you, you know, you're telling me all this great stuff- Yeah and now it's like, "And here's the quote." Exactly. It's like You know, I feel like the finance manager. My w- I, my wife just bought a car and, you know- Oh it's always, when you, when you do the r- thing and then you go to the finance manager, they- Right they're like, like, "Well,
how come it just cost me more?" Yeah. Which is always good. I'm doing the finance manager, like, "Hey, it's all this great stuff. You bought-" Yeah. " you know, we got you the, you know, the, the, the paint protection so you never have to wax your car or whatever." Exactly. So.
Oh my God, it's so crazy, isn't it? So here's, I mean, here's the thing, right? Like, there's a ton of different ways to approach this, right? Like, there's always gonna be a cost in migration, especially if you're migrating from an in- you know, an in- incumbent system that might be pure to a newer system. You know, there's, there's definitely sort of ways to approach this.
I think the way I like to think about is two ways. One, I think every organization needs to think about adopting an as-a-service model, right? Like, so, like, get, sort of get that concept of, of OPEX out, the concept of, "Oh, how much is the box going to cost?" And start thinking about, "All right, what's my cost of managing and supporting data?" And sort of move to that model, 'cause that's a
true, a absolute true thing. 'Cause you buy a box for future capacity, all right? Like, that's what you do. But when you as a, a delivered as a service, you're, you're, you're just managing the actual cost and effort of the data at a time, right?
So I think that gives you a ton more economic capability to really pay for exactly what you're using versus what you may use 12, 24 months down in the future. So I think that's a really good thing as well. And I think the other area where we need to look at is there is a significant saving on ROI and TCO over time, right?
So, you know, yes, you're gonna have an upfront, integration cost associated with it, but I think, you know, through reduction of operations, like there is all this studies around EDC right now that we're seeing between 70, 80% reduction in operation costs. I mean, that is h- huge. Like, the industry, what is the interest right now?
It's like eight people per petabyte of management or something like that, and Pure sort of comes in at, like, 1.5 or something. Like, it's incredible some of those things as well. So I think when you s- think about from a cost structure, you just sort of go, "All right. Let's move into an as-a-service model." Makes a ton of sense to be able to focus on paying
for the data management that we use versus the expected capacity in the future, and then taking advantage of all the operational efficiency and the reduction, like And I think then you, you have other things as well. Like, what about power, power, power and, cooling, right? Like, you know, like, I think we're like, we're at a 95% less footprint
for our capacity stuff. Mm-hmm. You know, there's things like that, that all creep into the cost aspects as well. I think it's sort of like cost is always one of those things. Yes, there's going to be an upfront cost of any migration, no matter what you do, whether-
Sure, sure you're expanding or whether you've got those things as well. But, like, really taking advantage of, you know, how do I reduce my energy bill through, you know, cooling and power reduction? How do I d- reduce my operational co- context through automation associated to that? How do I reduce my capacity per rack?
You know, how do I put more capacity in from a terabyte per rack? All these things sort of add up, and in the end, we're seeing most organizations, you know, have a reduction in cost between, like, 60 and 80%, depending on their workload stuff. Like, that is the ROI I'm looking at, right? Like, yes, if I'm gonna buy an EV- Right.
My initial cost, I've gotta go buy that EV, but over five years' time, three to five years' time, mate, I'm gonna be m- making money because I'm not paying five and a half, $5.30 I paid for premium fuel in my Jeep yesterday. $175 it cost me to fill my Jeep up yesterday, right? That is- I believe it. My actual- I wish my gas was that cheap.
We're- in Washington, we have some of the most expensive gas in the country. Yes. But we'll, we'll, we, we can move past that. Like, that sounds like a deal to me, so. All right, well, so, it I, I have to apologize, to the audience. Producer Laura just chimed into my ear.
Obviously, if you have any questions that you want Michael and I to answer, please throw them in the Q&A panel, throw them in the chat. We'll do our best to answer them live, or we'll, we will have time at the end. I should've covered that at the beginning. I'm so sorry. Again, I've got PTO brain, so I kinda missed
that step, so I apologize in advance or after the fact for that. All right, Michael. So we've done a little bit of questions you said to, but, like, now what does this actually look like? Like, we've had- Okay we had some conceptual layers and control planes, but, like, really, what does that look like, if you wanna walk us through the next piece of it?
All right. Let's look at this as well. So we have this thing called Evergreen Architecture, right? And that is sort of the way that we build our technology. Like, most people would know about our FlashBlade and our FlashArray, technology, as well as what we call now of the Pure Cloud, what's called Cloud Block Store.
It's essentially a m- mimicking That's, god, they'll kill me if engineering hear me say that. Of a FlashArray- I love this company or FlashBlade in Azure AWS or GCP, right? Yeah. So essentially it feels like a FlashArray or FlashBlade sitting in- Looks the same exactly. Looks- Yeah feels exactly the same.
And that is because our, our unique way that we think about Evergreen Architecture, right? Like, when we, when we built our systems, we were like, "Hey, you know what? This cloud stuff is really, really interesting the way that they abstract hardware from software." So we architected our systems the same. Like, if you think about our DFM, our version of SSDs, like, the We went, "Hey, you don't
need all that firmware components on the SSD. That all can be software. So let's make the SSD as simple as possible. It's NAND and with a translation layer, so that way we can grow it faster." Yeah. We have less energy things as well.
And then we put all these things in software, so we can abstract a ton of stuff in software. So that gives you the ability to have a couple things, right? Traditionally, a system is a block, a file, or an object system, and sometimes you have, like, block systems with file on top or file systems with block on top. There's problems with that as well, right?
What we do is we create a global pool of storage at a unified data plane level, so our, our FlashArray and FlashBlade systems sort of support n- native protocols. But, but since you build this sort of fleet management capability on top of it as well, I can natively support block, file, and object under a single operating environment, right? So I don't have multiple operating environments.
So I manage that exactly the same. I don't have, you know, file dependencies affecting my block- Yeah you know, deployment, right? Basically I, you know, I have that capability acro- across those things. The system determines whether that should sit on a FlashArray or a FlashBlade or Cloud Block
Store based on SLAs, based on what my capacity and repo- performance needs are. I don't need to make that decision anymore. So that comes because, again, we have a unique way we architect the hardware through DFMs and the arrays, and then we have a single operating model across FlashArray, FlashBlade, and, and Everpure Cloud, which we call Purity.
So that's the way you interact and manage those arrays. So n- you're managing sort of at that virtual layer from a unified data plane perspective as well. So that just gives you, again, what I'm, I'm now thinking about a global pool of storage where, where I have native protocol support, so I can just build my storage applications on
top of that without having to worry which physical architecture is underneath. Yeah. Exactly the way the cloud works, right? Exactly. Yeah, and that's And it's funny because we I'm gonna go back to the question of data lakes. Mm-hmm. Of, or, or again, the, the, the What, what happened historically, right? Like, if you rewind, is the thing that also
makes us unique, unique, as you've touched on, is what makes this even easier and possible now, is we use Purity, our common operating environment, across all of our solutions. Yeah. Where I go, I, I wind my clock back in my brain when I was managing storage from several different vendors- Mm-hmm at the time of data lakes, and even within their own product portfolio they had different- Absolutely operating environments, right? Yep.
So you had a E- even though they talked about a, a, a control plane, whatever, it was like another piece they had to stitch together because they had to talk OS one and OS two and OS three to make this work, where we've always had the same O- Operating Environment across our systems. Yeah. So they- Yeah we speak the same language- Mm-hmm and it just makes all this easier to do and easier to consume.
You're not having to, like There's not, like, a translation layer between, like, oh, now you're talking to OS one, so this means this. Exactly. Going to OS two, this means that. Exactly. So. Ex- And that makes things so much easier. Things like non-disruptive upgrades become real non-disruptive upgrades, right?
Like, things just happen a lot easier, right? And then I think once you have a unification of a data plane, once you can have a Once you have that virtual cloud of data, you can start doing amazing things with that, right? And I think that next thing is, like, you can now build a true global policy manag- engine on top of, on top of it, right?
Because that's the biggest issue right now. How do you apply global policies into fragmented environments? Mm-hmm. Like, it's super-duper hard to do. And, you know, you, like And that's where a lot of the operational friction sort of comes in. But now you have a unified environment, you
can now put a unified management layer on it. So I can now understand, like, I can now understand the entire capacity, utilization, and then apply a ton of stuff to it. Things like work row rebalancing becomes a dream now because we have a Purity operating environment, so we can move data from one environment to the other seamlessly without
any type of translation, any type of changing of that data. You can now sort of easily rebalance environments as new applications come in. You can enforce compliance, regulation, and governance. I can build my data taxonomy and say, "All right, you know, this data cannot appear in these volumes, and if that data tries to move to those volumes, it just won't happen." It
was like, "Mm, you can't do that from a compliance perspective." Bec- Because now you have visibility across all the arrays and all the- Mm-hmm physical locations of the arrays, as well as all the performance and SLA capacities of the arrays. So operations becomes easier. Data workflows become easier.
I can def- de- define global policy management so much easier. It, it just takes all the friction out of managing- your storage- Yeah which is the problem in the past, for sure. And one thing I would add to that, 'cause I would be very remiss, because it's something dear to my heart as working at Everpure, you mentioned, like, data locality, which is a big concern, right?
We all know about GDPR. Yeah. I always get those acronyms wrong and stuff. But the other thing to think about is we also pr- provide security layers to that. Like, if you're familiar with our SafeMode for, sort of for ransomware protection, right? 'Cause obviously securing the data is huge, so those policies are in there as well, right?
So you've got a kind of a end-to-end, lack of a better term, sorry, you know, to way to, like, data locality, data security, data sovereignty, like, all the things you talked about. So like, I just want to make sure we throw that in there as well as- Oh my gosh from our SafeMode, snapshots, replication. Like, again, all that's in here as well.
So it's kinda just think about how you're managing- Yeah and securing and protecting your workloads. And it's, it's funny you bring that up, 'cause I've been working a lot with the European team right now around data sovereignty and, and, like, and what a fascinating subject that is. Like, just- Yeah. Like, it's a fascinating subject.
But one- It's just fun to say sovereignty, quite honestly. Sovereignty. You know, it, it's a hard bloody word to spell, though. Like, I want If you keep it a secret, I made a macro that I t- when I type in capital S-O, it just w- writes out sovereignty, so I never have to type it wrong again. Yeah. So but one of the things is, like, one of the,
the things right now is they actually have, regulations about, the, reapplication of data after, after a, a, like, emergency. Like, so if data is corrupted or, or there's been a cyber resilience issue, or it's deleted and that data has to be restored, there is data sovereignty r- regulations associated with that as well. So this comes into that intelligent control
plane aspect, as you were saying. Like, when you have a full vision of replication, backup, and governance around your, like, your data security, you can reapply data so much faster now than- Yeah than ways of doing things in the past. So it's, it's super interest- And, like, it's a very, you know, flexible capability, but it
You can't have You know, like, people did meta management in the past. Meta managers just don't work. They work for a hot second before someone updates their integration module management. Yeah. We don't have that problem, right? This is just inherent in the system.
Pure1, Fusion all come together integrated into Purity. I mean, it's, I mean, Fusion's part of Purity anyway. It's fully integrated anyway. So there's some really good things here. But I wanna sort of talk about This is exciting, right?
Like, this is all Like, even those things are fantastic, right? But you talk about under the hood. We've talked about storing data and governing data. I'd like, I am super excited about the third layer that we just I've, I think I know where you're going. We just, we just Understanding
your data, right? Yeah, yeah, yeah. This is that, when we go back to the data primacy thing we talked before, this is the reason why you need an architectural model that can do this, because AI requires you to understand your data, right? And when you think about, like, why did L- like, there's two reasons.
One, why did we go from Pure Storage to Everpure? Because we're moving into the, the data intelligence. Right. Yeah, for those that don't know, we rebranded. Surprise! Like- We still love storage. Yeah. We're a manufacturer at heart, right? Yeah.
We still love storage, but this is Organizations want to be able to control their data from silicon to AI across the stack, and they don't want multiple vendors to do that. They wanna be able to support one vendor to be able to do that, and we were like, "Hey, let's do, let's get into this space." So we had that acquisition of 1touch, now Everpure Data Intelligence.
And so the great thing here is now you have a model where you can apply a l- intelligence da- data layer across all your environment and do discovery, classification, and map that. So now all your You understand. Like, a great example of this is, for example, like, I'm, da- I'll bring
up data sovereignty again. You know, so, you know, German has some really, strong data sovereignty regulations. Mm-hmm. So if you had PII German medical information sitting in a database in Brazil, for example, this system would find that because it understands the classification. It respects the taxonomy and say, "Hey, guys, you need to move this data back into a German
sovereign system." I'd, I don't do that. I'm a universal data intelligence layer. I can tell you it needs to be done, but then it goes and talks to the intelligent control plane, and the intelligent control plane says, "I know how to move that data, respecting your SLAs, respecting your compliance, respecting your governance.
I'll move th- that data, or the application, or the entire volume, move that into the data center, rebalance things around." I mean, this becomes beautiful. Now, the good thing about the intelligence data system, though, that I, I do wanna me- mention is it, because it's a data layer and understands all data, it doesn't actually require the intelligent control plane and unified data plane to work.
So you could, if you're running a non-Pure system, you can still do the univer- you do still do all this understanding of data. It will still spit out the report to say, "Hey, your c- stuff isn't in compliance right now because you're breaking data sovereignty rules or compliance rules or SLO things." It will be able to understand all that and give you the, the report to say, "Hey, you need to fix
this," and then you need to put it into your own operational context engine. Sure. The value for us is you don't need to manually do that. Like, the system is just incest- incestuous is probably not the right word. Yeah. Not, not a good word. It's integrated. But, but I get what, but I get what you're saying. What a fraud I am.
I want to re- The system is integrated- Yeah to be able to automate that, right? I wanna, I wanna repeat something that you Like, this layer from- Mm-hmm black cover storage agnostic is what you're telling me. It is 100- That's what I just heard. Yes. It's 100%- Okay storage agnostic.
So if you're running n- a non-Pure//E environment, right? Like, one of the big ones, one of the small ones, doesn't matter. Like- We won't mention names, but- No, exactly. Right. It, it, you can, you can now go and, discover all that data, map it, and classify it, and then apply your data taxonomy to that later to say, "All right,
here is all my taxonomy. Here's my regulations and requirements for data around sovereignty, around governance, around requirements, around performance, around SLOs." And it will report out to you. You build, like, agentic agents in it. It will report out all the d- data workflows to say where you are in compliance and where
you're not in compliance, and then you can just integrate that into your own operational context. So if you're running a non-Pure environment, you probably have some form of operational system hooked into a ticketing system. You can just do it that. As I said, for us- That intelligence da- data
layer talks to our intelligent control plane. So if you want data to move, like you can- Yeah, you get the automation it will say, "Hey Jason, I need to move this data. You okay with me doing that? Here's all the steps I'm gonna do." You go, "Yep," click.
Yeah. And then you go back to drinking your coffee, watching YouTube while it does it. Exactly. Got, got my Starbucks. Sorry. I support s- my wife, 35-year Starbucks employee, so it's, it's always Starbucks.
No, I, I'm- you know, that's- lived in Seattle for 20 years, I love Starbucks. Yeah. So that's great, Michael. And I feel like, like this is where you can tell me, I feel like this is where we could probably transition to the compare the approaches to what others are doing. Yeah. Or is that a fair statement?
I think so. I feel like you, you've set this table of, like, we're agnostic- but, like, what, what are we getting from that, you know, the other guys aren't doing? Is that fair? Yeah. Yeah. Yeah. Let's, let's do it. Laura, let's wind it up.
Let's see, we- put her on the spot, Producer Laura. All right. No worries. W- well, tell us about the approaches, Michael. Yeah. So I think the biggest thing here is, like, the, the, the challenge right now that most what most organizations are facing right now is, again, there is a fragmentation of the
way they've d- applied storage infrastructure across their organization, right? And so this is why I'm like, you know, a lo- a lot of people say to me, "Well, like, can I just do this with my current systems," right? And I'm like, "Well, you can do data intel. You can do classification, and you can do d- knowledge mapping and things like that in your
current systems." But again, like, i- it's going to tell you what is wrong or what's incorrect co- you know, compared to your c- your compliance and governance taxonomy, but it's not gonna help you fix it, right? You're still gonna manually have to go and do a bunch of stuff, right? And then from an AI perspective, again, this will tell you, this will
expose to AI all the data. Like, it will find all the data for you. And so your AI's like, "Yay, my data has now jumped 50%." That's not a good thing, right? Yeah. Like many ca- things, right, as well. So again, I think the biggest thing f- you know, like that we do differently that other
people don't do differently or other, other people don't do is, again, we started at a level of unification first, right? I think organizations that don't do this cannot architect their way out of it. They can't plug, plug in their way out of it. You know, they just can't add a new system, right?
That fundamentally, the bottom architecture, that first spit of C code they write, jeez, I'm dating myself now, right? Yeah. Is fundamentally different, right? So, you know, when we think about, like, what's sort of, sort of different as well, like, most organizations still work with, application storage silos.
We have a global pool of storage with native, native applications. Most organizations have multiple management systems to manage those as well. We have one unified intelligent control plane a- across the si- wa- way as well. Most organizations don't do fleet management. We have a single fleet management tool called Fusion.
If you're organizing, people out there are running Pure, well, stuff right now, you haven't turned Fusion on, turn Fusion on. It will change your life overnight. Even just the remote management. Right. I mean, right. Like, that, that's the- You
don't have to do all the stuff we were just talking about, but, uh- Even that, just that cool little thing just makes your life so easier. And now you can build blueprints and recipes and stuff like that. At the end of the day, like I like to think about the thing that we do, most storage, is like gardening. I hate gardening.
Like, I'm not a fan of it, but I used to live on this like two acre property back in, Texas, and it had 20, 30, 40, 50 different plants on it, and it was so annoying. I spent Sunday afternoons baking in that bloody heat- tending all these plants to make it look pretty, right? And I had to understand fertilization and, you know, what was I had to cut things at certain times.
It was annoying. So, and storage is like that today. Like every single, request is a new plant that you have to tend, feed, prune, right? All this type of stuff, and that's where all your time is going. There's no consistency in the way that we're doing things 'cause the system's forcing you to do that.
You wanna move from this gardening model to a farming model, where I offer crops. Here are the five crops you can consume from: high performance, high security, high reliability, repli- you know, whatever you gold, silver, bronze, whatever you wanna call those tiers, you can, you do those as well, and you consume from those tiers. You know what the great thing is? Now you're self-service.
I don't have to deal with you, I just have I monitor the health of those crops, and those c- those crops are then automated through a virtualized system. No other systems can do that on the planet outside of public cloud effectively for storage, except our ev- Pure platform, because we architected it like that 15 years ago. Right.
That's our uniqueness. Our evergreen architecture, 100%. Perfect. I love it. So Michael, I, we're getting close on time. Yeah. And I wanna be respectful of folks, so we're gonna swi- switch to some questions.
Sure. And I, we actually got a few from the audience, which I appreciate. Paul, I've seen you got a couple in here. Hopefully, we can answer these. I'll be quite honest, some of these might be a little bit over my skis, but we'll do our best. That's fine. So, but Michael, the first one we've got is
from Brian, and he asked I mean, this is, actually, this is a pretty easy question, or not easy question, but a great question, is, how long does discovery take? Like, is it based on petabytes? So we talk about like common pool and all this other stuff, and like seeing the data. Let me see if I can get the, get the number here.
Yeah. That's why I was like, might- There's a, there's a- might be a little in the weeds, but I do my best. We're gonna do our best, Brian. Yeah, yeah, yeah, yeah, yeah. So I think right now is like, so we do, en- entities. So we do a mapping of entities.
So it's not mapping of volumes, it's mapping of entities. And so right now, I don't know the speed of it. Like I, I, I don't, I, unfortunately, I can't tell you that. We'll have, we can find that out. Like, I'm assuming, like Jesse is gonna watch this one day and go, "Oh, you should know
that." But I can give you a, I can give you a couple of stats. So we do, we do a, a 98.6%, discovery and, classification accuracy. Through our 1touch technology, our EDI technology. So that's pretty much, like let's round that up to 100.
Like that's a good Well, why not? Yeah. Right? across 1.7 billion, entra- e- entities as well fro- from, from that, from that data perspective. So, you know, that's the data that I, that I know from now from a classification perspective as well. I do- again, I don't know from a speed
perspective- Sure how long that takes, but we can find out. Um- Yeah, what we can do, Paul, or I'm sorry, Brian, is we'll do a little digging. Because what my, my thought is, and maybe Brian you can answer this if you're still on, is like let's, let's assume Brian is an Everpure customer. I'm not saying you are, Brian, but let's assume you are, and he's like, "Dude, this
sounds great," and he flips the switch to like turn it on. You know- Yeah like how long is it gonna take? Like, let's say he's got, you know, maybe he's got five arrays, then like- Yeah. How long is it gonna take? So I'm assuming that's, might be where that's going with that. You know, he might have half a million, half a
million or a million or so entities, like how long is it gonna take to do- Yeah discovery? Like it's not years and months. Yeah. Like, you know, I think it's like minutes, hours, sort of like, you know, depending on the entity stuff. Like again, I'm g- I'm guessing, you know what I mean? Yeah.
So, but you know, from my, from my sort of knowledge in the, in, in talking to that team as well, I feel it's fairly rapid, from a discovery perspective. Yeah. And you know, and you know it does knowledge mapping first, and then does The classification takes of course a little bit longer because we're cracking open the content to have a look at it.
If- Yeah. So if we're, if we're on the right track there, Brian, which I hopefully we are, like let us know. We'll definitely follow up with you 'cause I feel like that's probably where it's like, hey, if I turn this on, day, week, year, minute? You know, what's it gonna take? Yeah.
So hopefully we're on the right track there. The other one, we've got a couple questions from Paul. Thanks, Paul. You probably have a good answer for this one actually, Michael. Cool. It says, "How does Everpure's approach compare to newer Data Ops and data-as-code trends regarding operationalizing
data management?" Which I feel like is gonna come into Fusion and your SLAs and all those stuff you kinda hit on. But I feel like- Yeah, yeah, yeah maybe just a little bit more- You know what? I think- of construct on that. Yeah, yeah, yeah. I think, and, and, and there is We have, we have capabilities like that too, right?
When you think about sort of the sort of modern, modern data, data ops, when you are starting to use a lot more AI agentic capability associated with that. And we fully integrate into that, and also third-party systems as well. So we have our own version of things like Copilot, which enables you to do sort of agentic agent, management and, and natural language interaction with that as well.
Plus a lot of our systems are from a, a going agent-based anyway. So you think about data workflows, like, EDI is a great example of a data workflow where you actually have, you know, like if I'm breaking an SLO, EDI might find that because I'm break- breaking that SLO. It will then, an agent will go and, you know, a codified agent will go and, go and research
that, and will actually then call up other agents to do further research or further fixing and stuff like that. So same models for sure as well. Now, the good thing about it is our open API means you don't have to use our models in that, in that aspect. You can actually integrate in other models.
I don't know exactly which ones we support. It's up on our website, you can go and look at those as well. But very similar approach and, and again, you can utilize that type of capability, within our, within our API as well. Yeah, Paul, for maybe, and again, I, since I don't know who is an existing or is not a
customer, right, so some folks could be doing research. Paul, if you look up stuff, around like Pure1 workflows, like we have a Under the covers of a lot of what Michael talked about, no offense Michael, is like there's these, these tools and these code things that drive a lot of this, you know, automation. So we have workflows for backup recovery and moving data and provisioning storage, and
that's kind of where all this, this is where the, the rubber hits the road. God, I'm full of cliches today, Michael. I'm so sorry. Like- It, it must be me. Yeah. But, so there's some stuff there that, that's below what we're talking about here is like at a high levels. Like there's a lot of stuff to, that drives
that intelligent control plane. Like I said- Wonderful our Fusion- And you can- integration workflows, that type of stuff. You can codify the whole thing, right? Like we- Yeah, API, all that stuff. Yeah like we hook into things like Ansible and all that.
Like you can, like you can build your own Like, you know, it's like again, we're fully open from an API interaction, in- interactivity. Yeah. And again, Paul, if there's that, that doesn't fully answer your question, throw something in chat. We can, we'll, we're more than willing to follow up with you- Oh, yeah with an email afterwards.
Paul has another question. Well, Paul, this one's a little tricky, and this is where I'm gonna put my old IT consultant hat on is, it depends. I'm sorry to do that to you. But the question, Michael, is when using Everpure, are there latency and resource factors when AI training workloads compete with operational applications on- Ah the same
infrastructure plane? And that's where I'm like the depends comes in. That sounds great. So Michael, do you wanna take a swag at it first? I'm, I'm happy to, but um- well it is, and it, it does depend, right? Right. But I think what you can do is like you can start moving into, you know, sort of
prioritization of workloads and models, right? 'Cause again, it comes down to like, you know, the good old Wi-Fi prioritization. You know, like I prioritize Zoom over bloody Xbox in my house. Well, not in this house, but keep going. Right. 'Cause you know, like I've been on Zooms and
Zooms has gone crash too, I think as well. But I think, you know, that, that, that comes down to that depends aspect. You have Y- you can train the system, to say, hey, there are certain You know, when we think about operate, operating workloads and the SLAs associated with workloads, and then the SLAs associated with model training, I've got all that capability to, to modify those,
those dials to be able to sort of ensure I'm not affecting my mission critical application SLAs, performance latency IO, because I'm doing advan- some advanced training model. You got some more stuff to add to that, Jason? Yeah, I do, and it, this is where I, and again I'm being a little cheeky where the, it depends, is because I mean, obviously every, regardless of vendor- every array has a,
there's only so much CPU and, you know, there's only so much resources in an array. Yeah. But we do have thi- like I'm gonna use FlashArray as an example with, you know, it's our primary It does block and file but, you know, it's, it's, you know, it's our scale Up box. I had to think about that for a second.
But, you know, there's only There's not, there's not infinite resources, but there's tools within the Purity operating environment for QOS that can make sure- Yes that certain workloads don't step over each other, and so on and so forth. But I mean, if you do have, you know, if you have a test system, let's say, that just completely runs wild, potentially it could step on production, so But, you know, that's
where it comes into, like, how, how our systems under the covers work with our QOS functionality- Yeah priorities, that type of stuff. And that's why I kinda say it depends, right? Like- Yeah there Could it happen? Yes, but that's no different than- But I think this is where, like, the new agentic agent
approach, works as well because now you've got, you've got these agents that are, are on all the time looking for stuff like that. So now you've got So if you have runner wa- runaway applications that are not respecting QOS, then an agent will find that- Yeah and immediately respond to that as well. 'Cause there's going to be, in, in operational environments as well, like s- we're going to
start relinquishing operational capability to the system over time. There are things that I will always have control over, right? Like- Yeah mission-critical applications and sensitive data- All right, Skynet. Exa- exactly. I'm like, I wanna be able to press the button yes or no. Yeah. Yeah.
But there are some stuff like, hey, if I've got a, an application that's running, go and lower the volume of that application for me. But do it automatically. Like, and let me know you, you've done it. But, you know, that's where I think we're g- starting to start seeing this AI intelligent appl- you know, application into sort of both the data workflow, the data context workflow,
and the operational context workflow. Yeah. Perfect. Sorry, I was responding. I'm trying to do chat, chat at the same time. You're doing really well multitasking, my friend. It's impressive. I, you know, it's, it comes with time, just
like- the best cats to answer. I don't see any other, questions. It does look like I didn't reali- you know what? See, the I'm on PTO, guys. My brain. I didn't even know we were doing a raffle.
Laura had to chime in. Yeah. We just threw a spoiler alert. Thank you, Producer Laura, for, like, keeping me on track. But it does look like we had a raffle, and congratulations to Michael D. You won the raffle. I don't even know what it is, I apologize.
But congratulations. Hopefully that's been a nice topper to your week. I'm assuming he's an awesome guy being called Michael, so congratulations. Oh, gosh. No, totally coincidence. I don't see any other questions.
I thank everybody. You did hear a Skynet reference, Timothy. I, my age is, like, full on display today. Michael, any closing- That's why, that's why we're saying please and thank you to AI right now, right? Y- you know what's so funny?
I, when I write my plo- my prompts, I do say, "Can you plea..." And I'm like, "What am I..." Like You know what? It doesn't, it doesn't care. Like We're, we're prepping for the future. Yeah. I blame I have a s- I had a Southern mother, so I have my please and my ma'ams- That's fantastic and my thank yous and sirs.
But any closing, unless any other questions come in, any closing thoughts, Michael, you wanna give to the audience? No, I think, you know, like, let's go back to that original things as well, like, you know, I, I have this, this, this quote. I think it was Einstein, actually. Oh, geez, I hope it was Einstein, when he says,
"You can't solve tomorrow's problems with today's tools." And I think that's a really great example of, hey, you know what? Like, the, the, AI's changing the world. We, we gotta, we've got, we just gotta Yes, it's changing the world as much as we wish- Daily.
It's not even like- As the, the Gen X in me doesn't want it to. It's changing the world, and it's- Yeah and I'm like, all right, so now we have to, we have to change our architecture to support it while also bringing the, what we, what's important to us along, along with the point. I think data primacy, the model we talked about before, is, is a great
way to sort of do that. EDC is our way to sort of implement that across the Everpure platform. I feel we're probably best suited in the industry to be able to take advantage of that more rapidly than what's happening in the industry today. So come check us out.
You know what I mean? We've got plenty of options for you to really go and just iron chef us, to really make us, make, make us prove it, make us validate. Yeah. Yeah. Yeah, if you're a, an existing customer, reach out to your account team. If you're potentially a customer, you're doing your own research, you know, reach out to your VAR of choice.
You know, they can get you in contact with the account team. Website, you know, everpuredata.com, tons of stuff. Michael, you'll see Michael all over the place. So it should be going to, like, our demos section where you can actually If you wanna go a little further, I know there are a couple more technical questions.
Like, we've got a site called Pure360, which is our, our demo hub. So, you know, beyond, like, what we've talked about, you can actually see some of the stuff in action. Like, you can see somebody go in and use Fusion, which is part of the control plane, to set up SLAs and provision things and do all that stuff. So if you, if you or your team wanna kinda see
more of the nuts and bolts, you know, everpuredata.com, go there, look for resources for demos. You can see all that stuff as well. And then finally Oh, there we go. If you're not familiar, we also have a digital community as well. I didn't have the link, so we got this QR code.
You know, go out there, ask questions. Again, we're Michael answers questions out there. I answer questions out there. Again, this is for, existing customers. Again, if you're just doing your research, like, go out there, ask questions. We actually have a pretty vibrant community of even existing customers that might chime in,
and you can say, "Hey, how do you guys do it?" Like, "I, I've been using Pure for X amount of time, and this is how I do it." Like, so again, this is, like, the, my statement of trust but verify. Like, you can see what we're saying, but go out and ax- actually ask people that are doing it.
So with that, Michael, happy Thursday. Thank you for coming on. This was great. In two weeks, for those that are interested, we're, we'll be doing another tech talk. This one is gonna be focusing on Oracle Database and how Everpure makes that better. And guess what? There will probably be a little EDC
conversation in there as well because that's just making things better, everybody's life easier. But until next time, again, Michael, thank you. For everybody that's joined us, thank you. Producer Laura, thank you as well.
And we'll see you all next time. Have a good week.
Michael Leworthy
Senior Director, Platform Strategy, Everpure
Jason Langer
Senior Technical Evangelist, Everpure
Infrastructure teams have always managed applications. AI requires them to manage data. That's a different challenge, and most platforms weren't built for it. The Enterprise Data Cloud architecture from Everpure bridges that gap, bringing data intelligence into the same platform infrastructure teams already manage, so infrastructure and data teams are finally working from a single, unified view.
In this session, we'll dig into what it means to manage a platform built for both operational performance and AI readiness, and the changes when infrastructure can finally gain visibility into its data, not just storage.
Key takeaways:
- Why the shift from application-centric to data-first infrastructure changes how platforms need to be managed
- How Everpure brings data intelligence into the operational layer without adding tools, teams, or complexity
- What shared visibility across infrastructure and data teams actually unlocks for AI initiatives
- How to manage a platform that serves every workload, from archive to AI
We Also Recommend...
Personalize for Me
My Updates
No updates at this time.