51:24 Webinar
Rewriting the Rules of Storage
Join us for this technical webinar to explore how Evergreen//One™ Adaptive Tiers breaks the traditional coupling between performance and capacity.
This webinar first aired on February 27, 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
Welcome. Welcome, everybody. We are excited to talk to you about rewriting the rules of storage. Yeah, that was a good video. That was a good video. We're still getting, getting used. I think every- everybody on the webinar is getting used to the new name, Everpure.
I thought that was a, that was a clever, a clever little video that we played there, so hopefully you enjoyed it as well. All right, so what are we gonna talk about today? Many of you already know about Evergreen//One, right? Evergreen//One is Pure's leading edge, very best STaaS product,
really different from the rest of the world. But today we wanna talk about what's new and what it can do for you. And we'll start with, Shall we do a quick, quick intro about yourself, Ben? I think it's, it's on- Yeah the sidebar somewhere, but, but just to, to, tell people who we are, right?
We're on the product side. We're the people that, that try to make this, useful and powerful for you all, so we appreciate you joining. We do, also appreciate questions or comments if you, if you have any at any point during the conversation.
Ben here has been with Pure for, gee, almost a decade, isn't it? Haven't you, Ben? Yep, yep. Working on every corner of the company through, through the history, through when, you know, it was Pure Storage all the way to now it being Everpure. I've been myself here seven years or so, and both of us are, are, some of the people
behind the scenes trying to make Evergreen One as powerful an offering as we can. That's right. That's right. I do wanna highlight, having been in the infrastructure space as a customer for a long time, a lot of what about today does highlight some of my own personal experiences remembering what it was like buying storage. So, thank you. Yeah.
Yeah, and I love that, thing where you're gonna take us into, Ben, starting with, with the old way. Evergreen//One is trying to be different, right, for our customers. It's trying to, to do something that is not It's not the status quo. It's not how storage used to be deployed, in a traditional environment.
So do you wanna start us off with where we're coming from, I think, as, as users and as purchasers? Yeah, yeah. You know, most of us are used to, the having a budget, right? We have, Every year we have a buying cycle and, you know, I remember I used to have to buy as much storage as I could afford, right?
Sure, I had some baseline idea of where I was gonna be, but the reality is you've gotta accommodate for growth. You've gotta take that, that risk out of the business, right? So you look at your budget and you tend to buy as much as you, you can afford in that particular budget cycle.
I don't know about you guys, but what that ended up, ends up looking like is a lot of times you're buying a box that's maybe more performant or specifically much larger in data footprint than you actually need, and you hope that you won't need any more, right? A lot of times you don't get another budget for storage for a long time. And so, what you do is you aim for the highest thing that you can afford, ask your vendors to
be as aggressive as possible, right? And, and then you buy what you can, you can get. And the reality is many of us are still buying storage this way, right? We don't have, a way out of emergencies in the future, right, other than buying more storage, and hoping that'll cover the next emergency.
And even when you look across the industry today at other STaaS vendors, it's still very much like this. You have to make a rough estimate of what might need over the course of the duration your contract and get a box delivered to your doorstep that, that can fulfill that need over time.
You got You kind of have to pick a box and hope that it's big enough, and then negotiate a price, and pay for as, as little of it as you possibly can, right? And that can be risky. You're at risk because you only have what's there, what's delivered in your contract under your negotiation to f- take on all of the potential workloads of your business.
And, and I think this is something that we do across not just storage, right, but as, as consumers, as purchasers in general. This is a very natural, however, however we wanna use the, the word legacy, it's a natural way for us to consume physical goods, right? When, when you go to the store, every time you go to the store, you're, you're trying to
figure out, "Okay, how much am I gonna need over some period of time? Should I buy the f- the, the small size or the family size, and how, how long is it gonna last me if I Or how much am I gonna use if I use it for the whole year?" the wrong choice, right? So I tend, typically, I tend to buy the bigger one, and then I see it in my, in my closet five years later, and it's
still half, half full. But, but I think it's a very natural way of, of thinking about products, about needs, about most things for us as consumers. However, when you look at what the cloud revolution, for lack of a better term, right, the cloud experience has done for us, it's also taught us that we don't have to do that
about everything, right? When you go into, into a new city these days, if you're, you're traveling, hopefully taking a little vacation, when you land airport, you're I don't think the first thing you're gonna do is go buy a car, right? You have a car at home, but, but you're not gonna try to forecast your car needs for the next 14 days, maybe 21 if, if we're all lucky, and, and buy a car
for those needs. You're gonna-Maybe take, take a taxi, an Uber, a, a Lyft, whatever it may be. You're gonna take a service that only you for what you're using because the only thing you're sure about in that short period is your immediate needs. You know that you gotta get from the airport to the hotel.
You don't know what's gonna necessarily happen later. You're, you're living it, you know, as it, as it comes. So you're just gonna take, take an Uber and be done with that. And I think that's We're seeing that time and time again, whether you see, look at cloud services, just to bring us back to the reality of, of storage and infrastructure.
If you look at cloud services, you're generally not asked for a five-year forecast when you go consume a volume on the cloud. When you sign up for a STaaS application, you're not buying as many seats as you're gonna need for the next five years, right? You start with, with what you have today.
And that is the experience that we as Pure are trying to, to bring people, even to something that feels as, as traditional or as commonplace as, you know, let me, let me just buy a box as big as I think I'm gonna need. We think it can be done a better way, right? So, so why don't we jump into that?
It really Again, I, I kind of have to use this, touching back on something you said earlier, Tago. I saw a joke, when there was a scarcity, right? I think there was some tariffs last year, and people were joking about paper they had bought during the pandemic, right? And so I, I still think of that, right, every time I think of, of overbuying and, and
potentially never needing it, so. And by the way, we should get back to that later because I think the, you know, the toilet paper that we all have left over, or some people, the, is a good indicator of what, what is the only solution we have in a time of scarcity to make sure we're gonna fi- be fine in the future. It wasn't the wrong thing to do to buy more
than you needed because you knew it was hard to come by, right? But would you have bought that amount if you had your corner store who gave you a guaranteed availability of toilet paper, who said, "Don't you worry. I guarantee that whenever you want another roll, it's gonna be here for you. And the price is not gonna go up," right?
So, so I think that's, that's what we're trying to do, not to make storage toilet paper, but- you get the point. Exactly. Exactly. So to that, to that example, right? What does Evergreen//One bring to the table?
It brings STaaS that scales as your data grows and doesn't have scarcity, doesn't have changing prices during the duration of your contract, doesn't have, you know, all of those things that, that we're just joking about with toilet paper. So tell me a little bit about how a scaling license works versus buying a giant box that you may never use. Well, well, I think what we are trying to
narrow down on, first and foremost, is where are you today, right? Where, where is your starting point? How much data do you have? What are you trying to do with that data? Which we then obviously, we're, we're technical people, we try to translate that
into performance parameters and whatnot, protocols, all of the good stuff that, that comes with, with the storage platform. But, we, we The, the key difference is that we don't wanna ask a question about We wanna really start with, with what you have today, and we wanna set, set you on an auto-scaling path, if you will.
This is what our traditional services do in Evergreen//One, right? We have different tiers of service. The different tiers are about how are you going to, how are we gonna keep you safe as your data grows? So if you double the amount of data that you have, you would expect the amount of performance you require to also scale up.
We don't know exactly by how much, again, because we're not trying to predict the future. We're focusing on today, but we keep you safe. We say if you double the data, you double the performance, right? Very straightforward. It is all taken care of by Pure. And again, one thing we should clarify for everybody, this is delivered within the four
walls of your data center, right? I know this all sounds like cloud services. This sounds like it has the, the, you know, all of the ingredients of, hey, have control. I, I no longer have it in my environment. What is it gonna do? I don't know.
Where is it stored? I don't know. None of that, right? Evergreen//One is trying to give you that agility by putting the, the work on us, but it's still fully delivered with appliances in, in your data center. It's, it's still pieces of hardware that you can see and, and touch and control for which
you have the key that we cannot access without your control and all, e- everything you expect, right? So let's just be very clear. The fact that this feels like an auto-scaling solution, but it feels virtually unlimited, I know we ran out of space in the chart here, goes to 500 TiB, but you can
draw those lines to, to infinity if you want, is not forcing you to be on the public cloud. It's still doing it fully in your data center, but with Pure Storage behind managing to that envelope. So, so maybe let's talk about that, and y- you can give us a, a, a better description. What does it mean to manage to that
That's a great question. So, you know, we have really low starting points, and one of the coolest things that we can do with Evergreen, as many of our existing customers knows, Evergreen as a platform allows us to grow as a customer needs to grow, right? If you're still buying arrays with Pure, that means that you can purchase a controller upgrade when you need more performance.
You can add a data pack or a shelf without ever taking downtime. All of those things can happen as you need it.In that old model, however, you're still having to buy that, and you still have to plan for that and think about that. One of the beauties of this platform is it scales for every TiB of data that you, you, increase your subscription.
You're automatically given additional performance, and what that means is starting small, you don't have to dedicate a whole rack to a storage platform, right? That has a huge amount of extra, you know, spinning disks or even FlashArray drives that you're never gonna use. We start with the right and appropriate configuration for your, consumption, right?
For your use case. On top of that, we don't even ask you to pay for the power and cooling of that, right? We give you We actually cut a check to your company to pay for the power and rack-and-stack space that our infrastructure takes in your environment, right? So that's a pretty cool starting point.
But what happens as you scale is it's never your responsibility to tell Pure Storage that you need more physical space, to tell Pure that you need more performance, because we watch that. We monitor all of that and automatically provide those infrastructure upgrades. Again, always non-disruptive because of Evergreen. We do all of that behind the scenes as your data size grows, right?
Again, you don't have to think about, "My goodness, my compliance officer came and told me I need to encrypt my databases," right? Now I've got to buy twice as much Flash. Evergreen//One takes that out, right? Stop thinking about that box that you used to have to buy, and really stay focused on, on the elements that are within your control, data size, and the amount
of performance you need. So that's a I think we've, Let's take a, a brief pause here. We've made the first leap. The first leap, and it's a leap that we've done as a company or we've made, years ago, right? We've had Evergreen//One or, and its, its former names, which we shall not name, for,
for many years now, almost a decade. The first leap is really going from purchasing appliances, purchasing boxes, future needs to sizing a service for the data you have today and letting Pure, Everpure in the background s- manage to that, to that envelope. Manage to the amount of data that you have, to the performance guarantees that Pure has put
in place. We manage to that, right? So the first leap is really moving from purchases to, to a service. But part of the, the point of this webinar, I think, is to not just get stuck at what we were doing, again, almost a decade ago, but show some of the things that we've done in the meantime that, that may, they push your thinking even f- even further, right?
It goes a little bit outside of, of what you would expect, us to be able to do. And that's what, what we hope to spend the rest of the webinar on. And, and again, please, if you have any questions or, or comments, please keep them coming. But what, what does this not maybe deliver the most value or, or work best for?
And where do we go from here? That's, that's a great question. So what about those workloads, right? So the scaling model is great. The scaling model's awesome for, for a lot of data sets. But what we find is every once in a while, a customer has, you know, maybe an,
a non-heterogeneous, data set, right? It's not all scaling the same amount of performance for every terabyte. It's not a, a single database or a single application. It's a bunch of different ba- applications. And today, maybe it fits on one of those scaling lines, but tomorrow somebody brings a,
a new demand, right? And that demand might be significantly more data size at the same performance level. Or in the case here, we have, a more performance, right, at a smaller in the old days, right, we would have potentially needed to upgrade a service tier and bring all of your data to a new service tier, or maybe even start a new subscription
to fit that sp- very specific need, right? But we found that not every customer can change, that not every customer can take on, new service infrastructure in their data center, right? So we, we needed to have a different solution, a solution that would give being able to scale in a direction that they need.
Yeah. And, and maybe to add to that, if we wanna go back is what, what are other solutions in the market for this type of thing? This is a place where a lot of our customers, tend to gravitate toward old-fashioned tiering, right? Old-fashioned hot tier, cold tier.
Right. They started trying to solve for this, solve for the workload that requires more performance, but maybe not as much data, by creating a separate tier of storage. This is how you end up with silos at the end of the day, right? You end up back where you started to some degree because now you're having to size
separate environments, separate tiers, each one independently, and the, the errors end up compounding. Yeah. Yeah, I, I remember back to the days, super large storage infrastructure, and you had storage pools, right? You had a RAID 10 storage data set, and you had a RAID 5 data set, and maybe you had, you
know, mirrored data set or something for different applications. Similarly today, right, we find people, you know, buying maybe, a, a different type of Flash for different use cases or being forced to buy from different vendors because pricing doesn't work from the vendor that they, you know, already have today, and you end up getting, you know, essentially data center sprawl, right?
Um-Uh, and, and don't get me wrong, right? None of that is wrong per se, right? Th- they're all the best solutions that we could come up with, with the tools we have, what we had at the time, right? When, when presented, you talked parameters and configuration in order to
achieve different performance, out of, different pools of storage. I mean, that's all we could do, right? That's, that was the only solution. What we're trying to present is not necessarily that, that those solutions were bad, but we are now in 2026, and there is a, we think, better way of doing that.
There, there really is. So talking to customers, reviewing their needs, and identifying areas where we could even do something better, right? Pure is really a software-driven hardware company, and the, the amazing thing that we have is the ability to, you know, pass on huge amounts of savings to our customers
based off of the value that we see across all of our customers. So dug through the data, looked at all of the use cases and all the different types of, you know, tiering that customers do, and why does s- this one customer have a small high-performance license and a maybe a large l- low performance license, you know? Why did they have to make that decision?
What, what benefits are they taking advantage of in order to do that? And then brought that all together, right, and made a s- a new service offering that is completely customizable that meets the needs of your independent or, or unique data sets, right? So we're gonna talk a little bit about Adaptive Tiers, but I wanna walk you
customer experience where this really ended up making a difference. They already had subscribed to one of our service levels. They loved Pure. It was working really, really wonderfully for them. They had, you know, a, a data set of about 200 terabytes.
Their legacy vendor, other arrays within their data center were coming up for renewal, and they knew that they wanted to go with Evergreen//One. They'd really enjoyed the uptime. They loved the fact that the arrays were installed quickly and, and their service started on a very specific date.
They loved the scalability and being able to add, add, capacity. So they really wanted to do the same thing for these other workloads. However, that service level they had if they just added that data to that service level, it wasn't gonna scale in the right way that they needed it, right? So their account team did a great job and worked up a, a solution that had a small
service a small license at a very high level to give them that database performance that they needed, and then, a very large license of a much cheaper, more archive-focused storage level. That would've now created the customer to have to have three separate service infrastructures that they would've had to manage and three licenses, right?
Again, not the end of the world. It's the way we used to do things. But in this case, they were able instead to say, "Hey, if I take this data bucket and this data bucket and then that high-performing database and bring it all together, what point on that performance-to-capacity curve will fit best for me?" And Pure was able
to dial in a service that gave them an e- the exact amount of performance they needed and not more, and the amount of data space that they needed. And that data that they were adding, especially that additional archival data, they didn't have to pay for extra performance just for that data, right? A scaling model, you're always paying, for every terabyte you, you consume, you're paying
for a little bit of performance and a little bit of space, right? So Taga, walk us through Adaptive and how that works. Well, look, it's, it's, it's funny because, looking at the, the comparison between the previous page and this one, n- now looks infinitely more complex, to me But what we're doing is giving you the knobs, right?
Think of this, it, it's a magical world in which you get to effectively create your own storage platform. You get to define it however you like. You're not, you're not bound to what service tiers or in the old world storage models, whatever you may want to think about.
You're not bound to what is on the catalog for my vendor, what is on the catalog for Everpure, Evergreen//One. It doesn't matter. It doesn't matter what's on our catalog. You get to define it yourself and only pay for exactly what you want to, to get. So now instead of saying, "Hey, pick a tier and then load your data," we give you the
controls, and the controls are effectively two. It's the s- the old one that we used to have, right? The how much data. That's super easy. Just load the data and, and you get billed predictably per terabyte. But now you get the performance control.
Now you get this additional knob or in this, in this view, what, you know, slider that you get to fine-tune to exactly the level you want. And most crucially, and this is where the adaptive part comes in, you get to adapt to your workload as your needs change, right? You no longer have a, you know, chain link, something connecting these two sliders so that
they move in unison. Now you get each one of the, the sliders separately, and the day a workload comes in demanding much higher performance but really, not adding a whole lot of terabytes to your environment, you get to do that. The day that w- another workload comes in and says, "I'm not gonna increase my performance, but I need to age my data another year.
I used to not do this, but now I do need to keep another year worth of data get you're gonna have to pay only for the data that you consume without asking us for any additional performance.This is not dissimilar when you think about it to the ca- the ability, the, the capabilities that you would have in a tiered environment, right? When you have hot and cold, you think about in, about it in these dimensions effectively.
You say, "When I want more hot data, increase my hot tier. When I want more Cloud Drive data, I need to age it, I just increase my cold tier." Well, here we've done away with the existence of tiers. We've given you the controls, define your own service, configure it however you want, and only pay for, for what you are trying to, to achieve, and change it over time.
A lot of concepts in there, Ben. There are a lot of concepts. Yeah. And, and genuinely, you know, many of our customers on the standard scaling levels, don't frequently find themselves in an, in a situation where they dramatically need one more of one than, than the other, right?
The scaling services are pretty well designed to match data types, right? Tier two storage, production databases, those kinds of things. But now you're talking about thinking about things in, in almost, a, a holistic view across all of your environment instead of having to have these targeted specific services.
Because there are times, exactly as you described, where a, a new application owner comes and says, you know, "Hey, I've got this, you know, petabyte of training data and, I just need to store it somewhere," right? Or maybe they bought some new GPUs and they need to, to, you know, get the data out of the system faster, but it's the same amount of data they've had before.
Now you can really di- directly control both of those, those situations. And one cool thing that this does is it gives you a very consistent and predictable amount of, you know, monthly cost for your performance element. You know if I need five more GB/s, I know how much I have to pay for that. I don't have to guess, you know, what kind of discount am I gonna get on that next
controller that I might have to buy. Do I have to buy a new array to be able to accommodate that extra workload, right? This is just, you know, literally just whatever you paid before and incrementing it up, right? Similarly with data, same type of STaaS in the, in the standard world, right?
If I'm on, one of the, one of our competitors' STaaS platforms, and I hit that limit of that physical box that I agreed to in my contract, right, I, I now have to sign a new contract that gives me more physical space. And again, I have to over-provision that space in order to account for future growth. With Pure, we don't have to worry about that, right?
Or let's say you own your arrays. You might have to buy a shelf. Which data packs do I buy, right? And trying to, again, figure out how much money do I have in the coffers, and how much space can I, you know, possibly get for it so I don't have to do this again soon.
But with Evergreen//One, it's just that's Pure's problem to make sure that you have that available to you, and you can easily increment one TiB at a time, right? One GB/s at a time on our block platform. So that's the, that's the type of flexibility that was never possible before because we always had to think of things in terms of blocks, right?
In terms of chunks of either infrastructure or contract size that And I, and I think, you made an important point there, Ben. The reality of what's happening under the covers, right? We, we all l- wanna know how this works, we all wanna know what happens. Yeah. Again, we, we said it very clearly, the This
is still delivered fully in your data center. It's delivered within your four walls. You got the, you got the lock and key, right? So under the covers, we are doing the work that you just talked about being the sizing of a potentially even an upgrade to the platform, sizing of, you know,
consolidation of modules, the expansion of blades in the case of FlashBlade controller, arrays on the, on the FlashArray side. All of that is still happening under the covers, right? What you are now effectively, you, you become the, the director, for lack of a better term, in, in, in an orchestra.
You're directing. You're directing where do you want this platform to go. Do you want it to go in one direction in terms of TiB, in the other direction in terms of performance? You get to direct, and we take all of the responsibility, work, cost, process, professional services, everything it takes to actually get the
infrastructure that is effectively part of your service upgraded, modified, configured in a way that matches your direction. Right? So it pulls you Frankly, what are we do with this is put you in the driver's seat from a direction perspective. You get to decide where it's trying to go, but leave the painful, work,
the commercial work, the quoting and the, the budgeting and everything else, the planning of where you're gonna be in the future, leave that out. Get a predictable rate. Leave the work to Pure. I love that. I love that. So this, it brings to mind, Tiago, you know,
I'm the infrastructure owner, right? I, I run my data center. You put this storage service in my data center, right? You're even paying me for the rack space and the power that you're consuming. Does that mean that I don't have the, the control of the infrastructure?
Does that mean I, my login gives me only the ability to, you know, create and delete volumes? Tell me a little bit more about that. No, that's the beauty. The beauty is that you have-All the control, all of the capabilities, all of the features, so the replication, the, the, all the newness,
even things like, you know, ActiveCluster, things that are more advanced, everything is fully included just the way you expect from Pure, right? There's no difference to that experience. The difference is who is going to do the work, who is going to do the planning, to make sure that what I was trying to get out of this platform truly is there.
Yeah. That's all that's moving, right? What's moving is that responsibility, that risk from the purchaser, if you will, to the vendor, us. I like that. I like that. 1touch other thing. So, you know, Pure's been talking a lot about Pure Fusion.
How does something like Adaptive interact with Pure Fusion? What, what kind of extra value can a customer realize from Evergreen//One, and maybe even specifically Adaptive when it comes to Fusion fleets? I mean, yeah, when you look at Evergreen//One, i- it's a great question, it's Evergreen//One and Fusion are, in a way, joined at the hip.
If you think about what Fusion is doing for our customers, it is allowing them to think about their infrastructure at a higher, more abstracted level without having to micromanage each appliance, right? It's giving the technical capabilities to treat a fleet of, of appliances of, and services and the like as one pool of data.
Evergreen//One is doing exactly the same from an operational perspective and from a deployment perspective and from a consumption perspective, for that pool of data. It's not just a matter of being able to manage many arrays as if they were one, but it's also being able to purchase, size, consume, do everything to the envelope of that data platform.
You're no longer having to configure to purchase in each underlying appliance. You're now defining the parameters of the envelope of the total data cloud that you're trying to deploy and having, Pure not only manage it technically with Fusion across different arrays and so on, but also deploy the infrastructure, on, a- as necessary to those parameters.
That's awesome. That's awesome. So you mentioned something that I think is worth bringing up, right? We are living right now in a bit of a, a world of when provisioning new infrastructure can be a real problem, right? Many of our customers already know that Evergreen//One, has a solution for that, and
it's a start date that is guaranteed from Pure, right? Can you tell us a little bit about that and why in this environment, in this, you know, climate that we're experiencing globally, why that's so important? Look, I'm gonna go back to the toilet paper we talked about earlier. And again, toilet paper isn't storage, but you can draw a parallel from even the, the
simplest of things. In times of scarcity or times of increasing prices or times of uncertainty, the only thing we n- we know will work as, as consumers is to overbuy, is to buy for everything you're gonna need over the next several years and then a little bit more just to be sure, right?
Running out is not an option because you may not be able to get any more of it, right? So you, you end up overbuying, you end up spending a lot of money today for something that maybe you didn't even need or, frankly, a price that may have come down in the future, right? So you end up taking on a lot of risk and, and you have to spend a lot of capital.
In Evergreen//One, yes, n- number one, we only, charge for what's being utilized, so we focus on the data today. But very crucially, we give you, every single contract, optionality, and that optionality shows up as committed pricing for, in this case, both performance and capacity increases that you may wanna do throughout the duration of your contract.
That contract can be many years, right? You can do a five-year contract if you want a price lock for that long. We have absolutely no problem with that. It is a way to get It's almost like an option, right? You get, you get the option of buying more at that price that you committed.
You get the option of expanding later, changing your requirements, directing us another way within a contract that you have already established. And we are contractually required to deliver within very, short 28 or 45-day depending on your, or your, geographical location, to those expansions, right? So you effectively have a vendor that, just like the corner store in my example
earlier, vendor is guaranteeing that there will be another roll of toilet paper when you need it so that you don't have to buy more today, right? And again, and I will stop with the toilet paper analogy now. But it's a powerful analogy, right? It's a powerful one. It's one we all can resonate with.
Now, the, it, the cool thing right now is we see some customers are struggling, right? Like you said, maybe they're, they're going out and, and trying to buy as much storage as they can, maybe different parts of their infrastructure. One of the great things about our Adaptive service, it's true for all of our services, but especially particular for our Adaptive service, is you can start really small, right?
You don't have to know today that you need 10 GB/s or that you need a petabyte of storage, right? The Adaptive service really lets you hone in and, and just get something in the door, lock in a unit rate for both the performance and the TiB capacity or the data size, that matches today's price profile.
So tomorrow, when AI explodes even more than it has today, right, and, and vendors are scrambling to, to raise their prices, you have that commitment, as Tago mentioned, right? You've got that-That commitment all the through the end of your term for that, that unit rate that you've subscribed to. And with Adaptive, that data rate is very, very fine-tuned for, you know, any type of
data, including archival data, because it's a very, very low price per TiB. And then again, that performance, only dial it up when you need it. And so you have this starting point that doesn't require a contract rewrite, it doesn't require an extension on the back end, and you can dial either one of those elements up at any point in time. Make your life easy, and you already know what
the price is gonna be. Exactly right. Ben, as we, as we start, you know, the end of the webinar, I would love to, still spend a little bit of time on, use cases, places where our customers have found tremendous value in thinking about their, their, data, their needs with the Adaptive hat on, with the, with the cloud-like mindset on. Yeah.
And I'm gonna start with one example that, that was super interesting to me. It was an example of, genomics researchers in, in healthcare. These were folks that, you can imagine the data set's extremely large. They do accumulate them. They don't end up deleting, genomes very often, right?
So these are massive data sets that they typically in, in a normal deployment would keep in a cold data lake, and then they copy over to First of all, those, those sets come from a sequencer, so they come from a machine that is producing the data, which has to operate on a hot type of tier. That data has to get copied over for storage to the, Cloud Drive, data layer or
data, pool, if you will. Mm-hmm. And then it gets rehydrated every time a researcher needs to touch 1touch of those, pieces of genome. So it goes all over the place. It There isn't so much of a time-bound component to it.
It is a question of what project are you working on. So what people end up doing is they end up copying massive amounts of data, everything related to the project that they're gonna work on, back to another hot tier, working on that, letting it age out. Mm-hmm. Part of what, what you're doing and, what I want to highlight as part of this
couple things. First, there is deduplication of data. So when you think about hot and cold, you are inherently deduplicating data because you don't delete it off of the Cloud Drive tier when you work on it on the hot tier. You're effectively caching it most of the time, at least in this workload. So the hot tier capacity that they was effectively incremental to the
data that they had in the Cloud Drive, in the Cloud Drive space. So now they had, you know, 10 petabytes of cold data, and then an additional petabyte of hot storage to make sure that they could manage their data set size, right? Mm-hmm. Mm-hmm. That was com- If they had been able to get the performance they needed out of the Cloud Drive, they would have only purchased 10
In this scenario, they had to purchase 11. And then number two, which is also extremely crucial, not only was this data oversized because of the deduplication, but the performance was oversized, and this is what was, super interesting. Because you're now not o- not only sizing in a tiered environment to the
workloads that are running off of the hot tier, you're also sizing to the copying, the transmission, the back and forth required from a throughput perspective and age out a, a piece of data, right? Mm-hmm. So you're oversizing the performance of the hot tier because now it has to do both copies and, workloads simultaneously.
And ultimately, for an end-to-end workflow, and this was the third are You have to take into account the like I said, from the sequencer to the lake back to a hot tier, in order to process the data, right? So your end-to-end workflow of each step of the way has also increased because of the copies that are required in order to work the data set.
Mm-hmm. Right? what does that look like in an adaptive world? In an adaptive world, like we said, it's a single tier. It's a single tier that goes as fast as you want it to go and is as big as you want it to, to be, right?
It gives you the controls. So rather than having a Cloud Drive they did is they sized the single tier's performance to the active workloads that they were gonna be running with their researchers. What that did is it effectively removed both of the copy jobs from the end-to-end process- Oh, wow so sped up the time from sequencing to actually being able to
and at the same time also decreased the amount of storage they had to purchase wasn't two separate things, it was one thing. So it delivered cost savings and end-to-end operational, time savings, right? Not to mention about the fact that now you don't have to think about the, the processes that are going to be moving data back and forth and the like.
That's awesome. Right? So it's, it's Again, we've have, I think as, as, storage people, we've spent a lot of time working with the tools we used to have, right? We spent a lot of time trying to figure out, Okay, given that I have hot and cold, how do I make this work? Yeah. Well, now we are trying to approach the, the problem from another angle, and what we
challenge all of our customers, to, to think about is: If I didn't have those constraints, how would I design the solution to be much, much better than what I was used to? That's a big challenge. It's the cloud thinking when you think about it, Ben. I think it's, it's, it's what we would expect when designing a service.
Again, not toilet paper analogy, but Uber analogy. It's what I expect when I go call a, a car and I wanna get from A to B. Um-I don't, I don't think about how many different cars and drivers are going to have to be involved and everything going on under the scenes. I just define the parameters and, and start at the solution I want.
Adaptive to me, as a, as a title- tagline, is an opportunity to start at the solution you want, right? To define that and then have Everpure, deal with the how. Well, and I think that's a, a, a very interesting use case, right? What would that, what would that customer have done in a different world?
Tiers. T- tiers. Tiers. They're managing those tiers, right? Yeah. Um- And again, and that's just one, one workload, one use case. We've seen several, whether it be where we started with Ben at my workload doesn't fit the parameters of your catalog, I wanna build my own.
Wonderful, wonderful use case. Whether it be a tier scenario where you want to speed up a workflow, whether it be another type of scenario where you want to increase your data over time, right, without keeping yourself to a, to a specific number. All of these are great use cases.
I, I encourage you to think in adaptive terms when you are trying to deploy a solution next time. So first think in Evergreen//One terms. Don't think about the box, think about the requirements that you wanna give us, and then direct us. Tell us what you want and, and leave the, the
challenge of how it's done and how to get it into a single system, a single environment, to us. I think that's, I think that's key, right? There are a few things in our own environments that we can really get to performance that our applications are currently driving, right? We might even know if the applications need more performance but they can't get it.
So that's, that's one element. And then typically it's pretty easy for us to know how much data we have. Might have to use a data collection tool, right? Pure's got, some, some data collection processes that we can help customers with. But ultimately, those are the two things that, you know, when it comes to my own data, those
are the two things that I need to know. How much do my applications currently store, and how fast do they need that data? And when you can hand that over to Pure and just take these little sliders and dial 'em up and over as you need it, I mean, it make, it can really make your life very simple, right? And when your, you know, database team comes and says, "I need five more GB per
second," you know exactly how much that's gonna cost. You don't have to worry about it. You don't have to source it. You don't have to put it out to bid, right? You don't have to worry about a, a forklift upgrade or, or in some cases even a controller
major change, right? You tell Pure that you need that change, and it happens. Absolutely. It's pretty meaningful, pretty powerful. It's awesome. Yep. Ben, any, any last thoughts?
I know we've, we've been talking, and we're very passionate about adaptive- Yeah, we are so we could keep going for hours. But- Oh, yeah for the interest of, of, you know, leaving time for, for our, listeners, Any last thoughts before we wrap up? No, it's just been a pleasure getting just a chance to talk about this.
I really appreciate ev- everybody who, who joined us today, and I hope that there's something here that you can take away and, you know, turn into a secret key for you in your business to make you successful, right? Make you be the hero in your own world. Yeah, absolutely.
And in all of this, again, if you ever have a conversation with your Everpure, team, please, you know, ask for this. They will, they will know what to do. But also play with it. It's available on our website. We have, super easy, you know, calculators and, and, sliders you can play around with to, to
kind of try this out and see how it evolves and what it looks like under So, please, please do take a look. With that, thank you so much for, for joining us today. I wish you all, a wonderful rest of your day and a wonderful weekend. Thank you, Tago. All right.
George has asked us a question. I'd love to answer that. In a nutshell, do you guys offer block, SMB, and NFS storage? Would you say that you are now competing directly with NetApp, for example, for both on-prem and NetApp CVO storage?
These are great questions. Thank you, George. So, all right. Block, SMB, and NFS, right? We have had FlashArray file as well as FlashBlade for many years now. But if you're looking for something that's all-in-one, that would be the FlashArray
platform, which we offer through Evergreen//One, right? That provides the world's best Flash for block, file, and SMB. So that can be NFS or, or SMB. We have recently been discussing and, and announcing, even further
enhancing the, unstructured data offering within that platform. So Purity FlashArray is now able to e- encompass virtually every type of data you would need, and that will be available through Evergreen//One as well. So now when we're talking about competing against I'm, I don't, I'm not gonna di- dive deep into, to NetApp specifically, but, competitors, right, many of them have,
really good strengths, and we are working every day to make sure that we can exceed those strengths. Tago, do you wanna talk a little bit about NetApp CVO and, and how that plays with our Everpure platform? Yeah, so, so when you look at Evergreen// One, it is a hybrid. We've talked a lot about on-prem.
A lot of our customers come from a world in which we're trying to get them from appliances to subscriptions, but, all of our offerings are hybrid. What a hybrid means is-It operates consistently both on-prem, and, on, on the, on the cloud. It can be done as a unified subscription, which spans both.
Your consumption can be aggregated between the two, so you can keep the ability to go back and forth between on-prem, and cloud as much as you'd like, move workloads one direction or the other. So all of that is fully supported and, and integrated into, into our products. Now, with Evergreen//One, what you get is that flexibility, right?
You get, you get the ability to move, to flex, to redeploy wherever you prefer and as your needs change. While obviously with traditional purchases, you're still able to access those capabilities, but you'll have to make a choice of where you want what. That's a great point. That's a great point.
And he further, further asked, "By CVE, I mean cloud storage via the usual suspect, AWS, Azure, and GCP," right? We can talk a little bit about the, Everpure cloud offerings. It's certainly grown in the past six months. Definitely worth reviewing and looking at many of our cloud-native integrations.
But Tago, any thoughts you wanna add there? No, it is, it's exactly what, what we just discussed about, right? Yeah. The, the ability to consume in a hybrid is there. You can also consume 100% on the cloud if that's your, your choice.
That will probably leave some of the benefits of Evergreen//One, aside, in don't necessarily have the flexibility to go on-prem. But for customers that are 100%, cloud, that is an option available to them. Awesome. George, I hope that answered your question well.
And as many of you know, Pure is always releasing updates to our Purity platform, so keep a, keep an eye out on all of the avenues of getting news from Pure, right? Go and look at our blogs. Check all, our news page frequently. I, I can't believe how quickly we are releasing new features and new functions, and
many of those things come from people like yourselves, providing feedback to us, just like the adaptive service did within Evergreen//One, right? Feedback, partner-shipping, creating a better product. Wonderful. Thank you, Ben, and thank you all for, for attending today.
Santiago Navonne
Director, Product Management, Everpure
Ben Lee
Principal Solutions Architect, Everpure
Modern enterprise workloads go from bursty AI and analytics jobs to unpredictable backup windows and evolving application estates. They demand an infrastructure that can flex without constant re-architecture, migrations, or overprovisioning.
Join us for this technical webinar to explore how Evergreen//One™ Adaptive Tiers breaks the traditional coupling between performance and capacity. Designed for existing Evergreen//One customers, learn how Adaptive Tiers enable you to independently scale IOPS, bandwidth, latency, and capacity—on demand and under SLA—as workload behavior changes.
Key takeaways:
- How Adaptive Tiers decouples performance and capacity to eliminate forced trade-offs
- Why non-disruptive, on-demand performance adjustments matter for modern workloads
- How a unified data space reduces pipeline complexity and operational overhead
- What Adaptive Tiers means for cost-efficiency, sustainability, and day-to-day operations
We Also Recommend...
Personalize for Me
My Updates
No updates at this time.