45:20 Webinar
RAMageddon: Navigating IT Infrastructure Volatility in Government
Explore how government agencies are tackling 'RAMageddon' and rising IT infrastructure costs with expert Don Poorman.
This webinar first aired on 8 September 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, everyone, and welcome to our session How to Plan for Change in Government IT. Hi, I'm Emily Jarvis, and I'm the senior manager for events here at GovLoop, and today I have the added pleasure of also being your moderator, so thanks so much for joining me. And a special thank you to our partners at Everpure for helping us put on this session today.
So let's get right into it, because AI is creating real momentum across government, but it's also accelerating supply and pricing pressures across the infrastructure agencies rely on. From memory to flash and high-capacity storage, agencies are facing rising costs, longer lead times, and more pressure on already constrained budgets.
This is being talked about as RAMageddon. These market conditions are not expected to ease quickly, and could continue into next year or the year beyond. So what can we do about it? Today our expert will break down what's driving the volatility, what it means for your
organization, and how we can respond. But before we get to all of that great content, I do have a few housekeeping items to go over. First things first, we do have some great resources available for you on the console under that resources button. This is also where you can find a little bit more out about our speaker, so make sure you
check out those learning opportunities. If you need access to that closed captioning feature, just look for that Show Captions icon along the bottom right-hand side of your screen, and that's where you'll find that added accessibility. Now, we hope you don't have any technical difficulties, but if you do, send us a message
through that Q&A tab, and one of my GovLoop colleagues will get you all sorted out. You can also use that same tab to submit questions, and we want your questions, so don't you dare be shy. You can also, of course, engage with your fellow attendees using that chat button along the bottom of your panel.
Finally, we are recording this training, so check your inboxes tomorrow, because we'll email you a link so you can watch the entire session on demand. All right. With all of that housekeeping out of the way, I want to introduce you to our speaker. Joining us today we have Don Poorman, who is the federal principal technology
strategist at Everpure. Don, thanks so much for being with us today. Thanks for having me, Emily. This is a fun topic. I can't wait to dive into it.
Maybe we'll make it a little less painful. Well, we'll, we'll try our best. Yeah. All right. So kick this off for us, because this is an enormous topic. What are the changes that you're seeing right now in this infrastructure
demand or pricing that's really impacting agencies and partners that you're working with? Yeah. It's impacting the whole world. Nobody is immune to what's going on right now, so I don't want anybody to think that we're all in some kind of unique boat.
No, we're all on the same ship right now as it relates to it. And what has happened is that we've had this perfect storm of a demand shock and a supply shock hitting at the same time, right? And in the middle of it all, we have federal agencies marching to the usual beat of the drum, which is you gotta do more.
You gotta do more AI, you gotta do more data, you gotta do more real-time analytics, with the underlying hardware, you know, needing to catch up to those demands, right? And what makes it worse is that hardware almost overnight became ridiculously expensive because of that convergence of those two shocks, right? And the toughest part about this is that the math changed while they were in
mid-procurement, and it's happening in real time. Any vendor that's out there right now will be telling you that, "This quote we're giving you is probably good for only 30 days," because that's how volatile the market is. And, you know, just as a general understanding of the landscape, your DRAM prices are up 171% year over year.
DDR spot prices have quadrupled since late 2025. That's only six months ago. And you gotta remember with this demand shock and this supply shock, and we'll talk a little bit about this later on, there are three manufacturers who control 95% of the DRAM production in the world.
It- There's a reason for it, so don't think this is some kind of conspiracy. This is just the nature of what happens when you have this kind of perfect storm. Yeah. I think you add in on the government side the budgeting process that happens for federal- I know agencies, and a 30-day window of a price to s- to s- Yeah, and they're like, "And w- we budgeted for this two years ago.
What happened?" I, yeah. Exactly. I, I understand. I truly do, and it's very, very tough to navigate. It is. All right. I referenced this phrase a, a minute ago- Yep but RAMageddon.
The Armageddon- It's a good name if you will, of, of this. We're, we're hearing this term kinda come about more and more. Yeah. What is it, and how is it showing up day to day for folks? Yeah, so to understand RAMageddon, as you call it, which I think is great, you have to
understand the nature of the RAM that we're talking about that is in crisis right now. And as you know, AI systems run on GPUs, graphic processes units, and those are throughput types of processors, right? So they need to move stuff quickly in parallel to keep feeding the processes that's looking for it, and the RAM that's associated with that needs to keep up at a similar pace, and
that stuff is called high bandwidth memory. And that is the stuff that everybody is retooling and building out right now in these foundries. Now, the high bandwidth memory gets manufactured in the same factories as the regular DRAM that you see in CPUs and servers and stuff like that, but they have to choose
what's gonna come down the assembly line, what is being bought more right now, what is the demand for here. And this is the thing, right? AI demand went vertical, GPU usage exploded, and everybody looked at the foundries to say, "We need high bandwidth manage- memory." So that's what's happened here is that
manufacturing has been driven by demand to specifically manufacture a certain type of memory. So when you say RAMageddon, it's not all RAM, it's a very specific type of RAM, which is different from what you see in servers with CPUs, 'cause that's latency-based. It's a different world, so you can't use both.
Y- it's not like I, "Hey, I can use my GPU RAM over here." No, you can't. That's just the way it goes. So, a- a- and then day to day for federal teams, obviously you're seeing it as price escalations on hardware quotes that they would- m- they might have been stable six months ago with, you know, some extended lead times and, and now you're seeing ridiculously
long extended lead times, "Hey, I hope the price is the same as what we said it was 30 days ago." And in some cases, vendors are forced to come back and renegotiate because we are at the mercy of the supply chain as well. Now, interestingly enough, okay, I'm gonna give you a really quick way back history story. Something like this happened in 2011, and there was monsoon flooding in Thailand,
where roughly 45% of the world's hard drives were manufactured. And that monsoon just took out, like, half of the global hard drive supply almost overnight, and prices tripled then as well, and it took two years to recover from that. So when you look at RAMageddon, you can understand now why it exists, right? Because of the rise of GPUs and AI and all that stuff.
But also know that it's kinda not a new thing, that we, we There are, there are very specific facilities that have to be available to build this stuff, and there are very few of them. They're very expensive to operate, which means they have to balance their production to what the demand is, right? And, and in the case of RAMageddon, those three companies, which are Samsung, SK Hynix,
and Micron, they control 95% of the global DRAM production. And again, it's not because it's a conspiracy, it's because these foundries are extremely expensive to operate, and there's risk involved in doing multiples, especially when you don't know what the supply chain is gonna look like further on down the line. Like right now, we got a lot of people who want it for the AI stuff, but that's gonna
wane at some point. You don't wanna have excess production capacity on the other side when things start to slow down. So this choke point, again, this isn't a conspiracy. This is just the nature of the industry with a very specific need as far as manufacturing is concerned, kinda, kinda saying, "Hold on, guys, I can only do so much here," right?
So, that's, that's the way to think of RAMageddon. Yeah. And it, it's not new. Like I said, this happened in Thailand in two- in 2011. It, it has me thinking about, like, a, a kitchen, and you're- Yeah you're trying to put out as much food as you possibly can. Right. Right, if you're a parent, if you're a parent, you got, like You got kids who want pancakes,
and you got kids who want chicken. You're like, "I can't do 'em both at the same time. Which kid do I love more?" You know? That's, that's kinda, that's kinda the way you gotta look at it, right? So- Yeah and again, we're all in the same boat.
If you talk to any OEM manufacturer, I'm assuming they will say, "Look, we're on the working end of this just like you are, so we're doing the best we can." Yeah. All right. So you already hinted at this a little bit, but I, I wanna get us specific here. Agencies have, like we've already talked about, a lot of different priorities and a lot of
different pressures. Where are you seeing them kind of taxed the most? Is it the cost piece, the supply chain, the, the planning uncertainty? We have no idea what's ahead of us. Yeah. Where, where do you see that tie being pulled the tightest?
So I'm gonna sound a little wishy-washy here. I'm gonna say it's really across all three vectors. But I, I wanna think that planning uncertainty is probably the one that does the most damage because it compounds over time, right? Analysis paralysis, delaying things.
Those things just compound the problems on the other side when, when the water pipe eventually opens and you're able to drink water again, right? And i- if I know something costs more, I, I could adjust the request, but if there's a delay, I, I don't know how to build a buffer in. I, I don't You know, there's a lot of risk and uncertainty in all of this, and if I can't
reliably predict what something'll cost six months from now, I, I, I'm not sure how I even manage that planning downstream, quite honestly. And what I'm seeing agencies that are building their fiscal year '27 capital plans is they're having to re-justify in-flight things that they assumed six months ago, if you can believe it, right? And that's expensive in ways that don't show
up in the budget line. It, it's program manager time. It's contract modification overhead. It's mission delay as well, which is a very, very thing that hits near and dear to me. I used to be in the Air Force, so this stuff really, really matters to me, and this
volatility just keeps people away from being able to make- trusted decisions that can last longer than 30 days, if you can believe it, right? I feel like, and you've hinted at this already, this, this change has come so quickly, right? Yes. You know, you've mentioned six months, 30 days. Yeah. How has budget constraint shaping IT decisions,
how has that transformed over, let's just, let's just box it 12 months. If I was budget planning at this time last year- It has to be 12 months, right? Given what's going on, yeah, absolutely what does the difference look like between 2025 and now in 2026? Yeah. It's, the frame has shifted from optimization
to optionality. You know, a year ago the budget conversation was largely about getting the best price on a known quantity, and now the conversation is about what happens if that known quantity changes, right? Because it's such a moving target. And the agencies that are asking that question are the ones who are building flexibility into
their procurement plans. That's the biggest one, right? They're looking at different consumption models, shorter commitment windows, mix-and-match procurement approaches. A- And the ones that locked in on the five-year capital commitments on fixed price
hardware assumptions, they're the ones that are having the harder conversations right now. And that's no fault to anybody, quite honestly, because many organizations are all about that fixed rate capital cost over a five-year period of time. That's what the industry generally expects. But the ones who've said, "Look, I get that.
I understand that, but the market is forcing me to think differently," which means I've got to start thinking about different consumption models and flexibility in my budget, they're the ones that are moving forward a lot faster, I think, or having more dynamic conversations. Not that folks with the five-year capital stuff are wrong, but they may wanna consider changing their tack at some point.
Makes a whole lot of sense. All right. I wanna pivot here slightly, 'cause I think we've spent a little bit of time already talking about some of the challenges. I wanna now pivot to talking about how we prioritize and, and some better decision-making, understanding the realities of the world that, that
we've kind of laid out here. Sure. And as you've been talking, it's making me feel anxious, 'cause I'm like, "There is so much volatility." I- I'm sorry. No, not your fault. Not your fault. You're outlining the reality of the world. But when everything feels urgent- Yeah like it is, right?
Yeah. It can often seem that everything is a five-alarm fire. So how do you, when you're working with your government partners, how do you help them decide which projects move forward, which ones get paused, maybe which ones get delayed, as the world is kind of figuring out what this new normal is, at least for today?
Yeah. And from my perspective, I'm going to speak selfishly, because I'm in the data management business, right? It- others may talk about computing for hypervisors or AI or, or what have you. I'm most interested in the data management side of things. And, and the honest answer is that the prioritization discipline is revealing that
which agencies have a mature data government pos- da- da- data governance posture, and the ones who don't. Because if you have clear visibility into what your data is and the missions it supports and where it needs to be and where the bottlenecks are, that gives you better insight into prioritization. And this is why Everpure has pivoted a, a little bit away, well, mostly away, from being
just an enterprise storage company to being a data management company. Because for the longest time we believed it was all about just storage infrastructure, right? We, we sold you the crates. And it's important to understand now what are in those crates, because you suddenly have to
make decisions that you can't buy more crates. You're gonna have to figure out what's in the crates and better arrange them in the crates. And that's why data management and data governance is such an important part of the refresh cycle now, because you can't just throw in all flavor to everybody with the data and just say, "I need to go from this to this.
I'll just lump it all together and move it." You need to really understand what comprises your data, what it supports, and what workloads are there. So obviously, visibility and an understanding of your data landscape is what I'm going to say from an Everpure perspective matters in working through your prioritization challenges. It, it's somewhat shocking to me still every single day about just how much data agencies
are producing and- Yeah have access to on a- Oh, yeah daily basis. Oh, yeah. In, in the 10 years I've been covering this, it has gone exponentially faster and more volume than I think anyone previously would have ever imagined, and obviously that's going to continue to grow. So if you don't have a handle on your data governance today, it's not
gonna get better tomorrow. No, and, and, and I think it, and it brings up a very interesting point, right? You've talked about this data explosion. Well, why is there a data explosion? Well, most of it's in the unstructured space- Right which is file shares, object storage,
things along those lines. Well, why is it growing? A lot of it is machine-generated, right? In the nature of AI and the nature of some of the stuff we do out there now, more machines are creating data than humans, right?
So this is where we all get worried that Arnold Schwarzenegger and Terminator's gonna come get us and whatnot, 'cause they all ha- they all have the data now. But that's what we're seeing, right, is that this explosion of data is, is coming. It's machine-based, and everybody is struggling with the groundswell, like you talked about. And- What was like even five years ago for
data management is completely different now. So, and again, this is why I go back to data government, governance, and visibility. Because if you are ever considering putting AI into your operations, you really have to understand your data. You really have to understand the governance, the nature of its classification, whether or
not it's allowed to be a part of a RAG pipeline, whether or not it's allowed to be a part of a model that's trained. And all of that starts with visibility into it and understanding it. Makes a whole lot of sense. All right, let's talk trade-offs. Yeah.
Because if, if we wanted to- Yeah I think in our dream world, we'd, we'd have it all. But budget alone is gonna dictate that that's not always gonna be possible. So what trade-offs are agencies having to make- Yeah to stay within a budget while still getting these mission needs accomplished, which we know is essential, and critically important here? What do some of those trade-offs look like?
Yeah, well, in a perfect world, like you said, we'd also all get a pony, right? Yes. That's, that's what we're aiming for here. Exactly. Unfortunately, we don't get ponies in this situation. But, I mean, the most common trade-off I'm seeing is a refresh deferral, right?
Equipment that was scheduled for replacement is getting extended, sometimes two or three years. And that's an understandable short-term decision, but it's a real long-term risk because you're kind of kicking the can down the road and converting a capital item into an operational liability, right?
Because you're just paying for maintenance on it. You're not buying anything anymore. And that's eventually gonna surface as an emergency replacement or a security incident, right? Mm-hmm. Because vendors need you to move on to the more current stuff, so pushing it out further
and further puts strain on you operationally, obviously, to do that. And the agencies that are, they're doing the smarter version of that trade are the ones who are restructuring how they consume the infrastructure, right? We talked about that earlier. Those who are instead of looking at a five-year capital type of investment refresh
kind of thing, they've moved towards consumption-based models where they're paying for what they use and not holding the depreciation risk of a bunch of stuff that they're not storing anything on it, right? So, you know, when we look at the constraint and somebody says, "Well, I wanna push out retaining my hardware two or three years," what I've been saying is maybe you can
reconsider how you provision the newer stuff, right? That And, and Pure has that with, with our Evergreen//One storage-as-a-service. You are able to just commit to an amount of storage that's got a set amount of performance and a length of contract. And it's up to us to make sure that we deliver that storage to you on-premises.
So again, it, it, it, the constraint is obviously financially driven, and changing how you think about procurement is probably the best way to lower the risk a little bit. Makes sense. All right, let's talk about procurement and finance teams specifically here. Sure. And we know historically this is a group, that loves change, and to do things differently.
Oh, yeah, the money people. Oh, the money people love that stuff, yeah. They love an adjustment. I love the finance people, by the way. Me too. I, I feel bad for them right now.
I'm married to one. I'm married to one. I can I, I get it. I get it. Procurement teams and finance teams, though, fundamentally in this environment have to adjust their approach. Yeah. What ways are you seeing that adjustment take place?
Yeah, it's, it's very interesting, right? 'Cause it, it's funny, we're in the middle of a thunderstorm trying to get an umbrella open- Yeah is the way I feel right now with everything. And, you know, like you said, the finance people, they, they don't wanna open up the umbrella sometimes.
So the procurement teams I'm watching that are making real progress are the ones that are treating the price volatility as a first-class risk category, right? When you go to procure something, risk management comes into play, right? Is the vendor good? Is the product good? Is the solution good?
And then the price kind of becomes the icing on the cake in many cases, because pricing wasn't as volatile before, and the government has fixed rates and everything like that. But volatility has now become a first-class risk category, and that changes the conversation from what's the lowest price I can lock in today to what's the most flexibility I can preserve given how uncertain the next 18 to 24 months looks, right?
And, and that shift usually means being more open to that consumption-based contract and more willing to use shorter commitment windows with those contracts at maybe a slightly higher unit price. Mm-hmm. But more aggre- it, it makes it more aggressive about, you know, building the escalation clauses into multi-year agreements.
So you, you may buy a lar- you know, a more expensive unit now. But in the long run, given what volatility is gonna show for the next 24 months, it'll be the smarter play over a much longer period of time if you consider consumption-based procurement, if that makes sense. That makes total sense.
All right, one of the things, that I know folks who are listening to sessions like this one, we can get a little lost in the sauce, right? Yeah. Talking about- Of course, yeah what this means, what it looks like. So I'm curious if you have any examples, and we don't need to name names here, but any examples of agencies that you're working with or that you've seen getting creative to really
extend the life of their existing infrastructure? Yeah. So I, I have an example that we can cover later on with it that will kind of close the loop on this. But as far as the agencies that we're working with con- are concerned is that the, the agencies are getting more deliberate about what they refresh in what order, right?
We're prioritizing is all this really is. And, and w- instead of treating infrastructure as just this uniform fleet that all gets turned over at the same time, every agency I'm talking to right now is like, "We need a hierarchy of workloads. We need a hierarchy of the storage that supports those workloads.
We need a hierarchy of the data governance that supports the storage that supports those workloads." So a lot of things that I'm seeing across all agencies, and we can talk about a specific one later on, is, you know, compliance data, archival workloads, less frequently accessed operational data. That gets in line behind the important stuff, right?
And they may extend the life of some of the stuff that that stuff lays on, and they're simply implementing for the mission-critical stuff, to include AI if they're bringing it into their environment. And I think the agencies that can make that call and the, you know, they made some prior investment in understanding what they actually have, those are the ones that won't
over-refresh or get stuck in a buying cycle that overprices them or something like that. And again, because I'm a data management guy, I'm going to go back to saying, listen, to be able to do that, you gotta have visibility into it. Mm-hmm. You've gotta understand what you're storing because you're right, you can't throw all of the infrastructure out anymore and bring the new infrastructure in.
The pricing will just make the finance people go crazy, right? So I think this visibility thing that we are definitely talking about with our agencies is, is definitely resonating well with them. Especially e- especially when it shows a real impact on the other side for what they need to pro- plan for budget-wise.
Makes sense. So obviously, what we're talking about here in a lot of cases are consumption-based models- Yep or cloud models. Where are you seeing them, be less, let's say, talked about or theorized- Yeah, yeah and where are they more being put into practice? Yeah. It, it, it's, it's a very, very interesting
time, right? Because it almost feels like, and, and I don't wanna brag too much, but- No with our Evergreen//One model, our STaaS model, which we've had forever, suddenly it's become the thing, right, that people are looking at. And, you know, the place where I see consumption models delivering real tangible value, and not just in theory, is decoupling
the timing of when do I need the capacity from when do I need to commit the budget and place the order, right? Traditional CapEx procurement chains, they smash those two things together. You have to predict your capacity over the next 18 months, commit the budget, place the order, and then just living with whatever the market did in between the planning and the
delivery date, right? Which that doesn't work very well right now Nope when you blink and everything went up 40% overnight, right? It, w- it's workable. That model I gave you is workable in a stable environment, but it's a liability now as far as that kind of planning cycle is concerned.
So when we look at Evergreen//One, like I said, our STaaS that we commit to the SLA, the capacity, and the availability of the storage. It's all on us to maintain. The agencies are using it specifically as a risk mitigation tool. You know, if we can subscribe to what we need now, ride out the 24 months,
36 months, however long it's gonna be, we can then normalize back to or evolve our procurement to a much more modern way of looking at it, right? They're, they're going, with, with Evergreen//One stuff, they're going consumption because it removes the single biggest source of planning uncertainty from their infrastructure budget, which is the hardware price assumption, right?
If you just subscribe to us in the amount of storage that you need, it's up to us to make sure that you have that capacity, that performance, and your price is fixed. That's the best part about it. You know, you're not writing a budget justification for a cost number that went up all of a sudden you had no control over, right?
We, we help ease the shock of that with a subscription model. And the best part about it is, is when you talk about that kind of a sc- subscription to mitigate traditional capital expenditure things, w- we take the exposure as Everpure, right? We own the equipment. We own the service.
You are just the end consumer, which leaves it up to us to make sure that the math works, obviously, which we got a lot of smart math people. So- you know, I, I think it's a risk management conversation as it relates to the consumption model or the cloud model or whatever you wanna call it, but the mechanics of it make sense when you look at the fact that within the next 18 to 36 months we aren't
gonna know what the pricing is a month from now, quite honestly. So as you mentioned here a, a little bit ago, we obviously have legacy systems. There are still many on-prem investments - Yeah, yeah happening at organizations. You've also got the cloud. You've got the AI of it all coming into, to the mix.
What are some ways or some strategies you see agencies balancing this portfolio with all- Yeah of these options? What, what are you seeing as some of the best practices there? So it is like plate spinning, right? I, I, I feel really badly for, for everybody in procurement as well as infrastructure
management and data management right now because nobody knows what's gonna happen 60 days from now. But I- my take on this is gonna be a little contrarian because- Okay you know, we've, we've pushed cloud for the last decade, right? I, I was there when we started talking cloud at another company. And, but I'm an advocate, especially when it comes to federal agencies, I'm an advocate for
data sovereignty, which means you have exclusive control over your production data, right? And when, when I think of hybrid cloud, while it's genuinely attractive as a relief valve for cost and capacity pressure, I don't think it, it really needs to be looked at from a risk management decision as to a long-term solution.
And because the moment your production data moves into a shared cloud environment, you've made a sovereignty trade, right? You've accepted that someone else's infrastructure, someone else's security posture, potentially someone else's jurisdiction exposure is now a part of your data story. And you've handed infrastructure management to
a provider who just sees you as one of a couple thousand customers, right? And while it's important to you, eh, you're one of 1,000 to them kind of thing. So I, I don't think there's a simple trade available, like let's all put it in the cloud and everything will be fine- Right right? Because you really do trade off a lot of stuff when you do that.
And it, you know, it bears itself on mission requirements, classification levels, and compliance obligations, and all of that. But, I'm not saying don't do it, but I'm saying price it correctly. So if you do look to pivot to leveraging the cloud for some of this, you need to build the sovereignty part into your risk management framework.
And it would be the same as putting in vendor risk or security risk, quite honestly, that, you know, what data are you moving? Who controls access to it? What happens to your operational posture if the relationship changes? Mm-hmm. What if Microsoft gets
out of the Azure business? They won't, don't get me wrong, but- you know, it, but those things are out of your control, and they can happen when you hand the keys of your data over to somebody else. And I think the agencies that I've talked to are kind of figuring it out and getting it right are the ones that are treating hybrid cloud as kind of a deliberate bounded decision
for specific workloads, but not general posture shift. Because moving everything into somebody else's apartment building, now you're at the mercy of the landlord as opposed to you- Right owning what you have. So you've got to be really careful with that. And again, I'm not anti-cloud, I'm not anti-hybrid cloud, but I'm You got to be smart
about it because it's not the silver bullet that many people want you to think, if that makes sense. It does. It does. And those are some really real concerns, and we've definitely seen where it's gone wrong, uh- Yeah in, in some cases. So, I think that's a, a really pragmatic, approach that, that you're talking about.
All right, we are rapidly going through this session today, Don. I just want to remind our audience that if you want to, you are more than welcome to ask any questions. But I got a few more teed up for you. And this first one, Don, is really around the idea of And it's a little bit
of what we've kind of talked about already, but there's volatility. We- Yeah there's cost volatility, supply chain volatility. There's so many things happening. Yep. So many feel out of control for agencies. I can't do X, Y, and Z about this.
Right. From their perspective, if you're sitting in, in an agency shop here, what are a few practical steps that you can take to say, "Here are what I can control to better manage this volatility over the next year to two years"? Yeah. So I'm a lifelong student of stoicism as
philo- Okay as a philosophy, which the core tenet of that is you don't control most of what happens around you. You just can only react, right? You have a very limited span of control- Mm-hmm which welcome to my world, everybody- because we have no control over what's going on right now.
But we can do certain things that help our situation a lot better. And I'll, I'm gonna give you a real world example, 'cause I, I teed it up earlier, right? I was with a customer earlier this week who's inherited 50 petabytes of data that they've been tasked with modernizing. Now, what modernizing means right now is, is still up in the air.
And, and the instinct with this is to just jump into the infrastructure question and say, "Okay, how do we consolidate? What do we refresh? What do we move to make this 50 petabytes easier to do?" But our position is, look, it's 50 petabytes of stuff you don't know what it is, and you just can't, especially with the market the way it is right
now, you just can't wholesale move it from one place to the other. Which I think is an interesting conversation to have, because what we're telling somebody, a customer, is we'll sell you the storage all day long. If you want 50 petabytes, we'll give it to you. But we also think that before you do that, you should go through an exercise to at least
understand what you have, classify it, understand its relationships, and determine do you need all 50 petabytes out of the gate, or do you subscribe to 25, let's say, through this bumpy part of procurement problems, move the 25 terabytes of important stuff over, and figure out a way to handle the other 25 until we get to a point where budgets are a little bit more aligned with what the prices of the market are.
So, you know, what we said was, it- we have a product at EverPeer called Data Intelligence, and what Data Intelligence is, is a software product that deploys and actually can read into the data at all of your locations. And when I say read into the data, I don't mean, like, tell me how many Microsoft Word files I have. What it means is if I go into a place and you
see a Don Poorman resume.doc, you can't assume that that's my resume. You could go into it and realize, oh my gosh, this is a list of credit cards that he's stolen, right? Not that I steal credit cards, but, but that's what we're saying here, right? Is we're saying, listen, for you to understand what 50 petabytes really, really means as far
as modernization is concerned, you have to understand the data- That is comprising that 50 petabytes, because if you skip to the modernization without knowing what it is you're dealing with, you're probably gonna find that the end answer doesn't completely align with what your vision was at the beginning of it all, right? And this isn't a technology recommendation for its own sake.
What we're saying is the prerequisite, especially in the world of data management, every other cost management decision is gonna pivot off of you understanding the data itself, right? And then from there, once you understand the data, obviously you could build your procurement flexibility into it, you can use flexibility in, in the architecture and, and
things along those lines with us to move what you need to. But it gives you a much more graduated way of doing things without jumping all into 50 petabytes, and really starting with, "What can I manage right now with the way the economy is?" Versus, you know, going all in on something that may not even hit the mark. And, you know, th- that kind of decision a- actually to me should be normalized from here
on out. This, this is a great inflection point for us as far as data management's concerned, because you do need to know what you've got. And before you do anything, before you turn wrench one, please be sure you know what you're storing, where it needs to be, and what it services. And that's, that's what we did with this customer.
And, and they're like, "Yeah, we- we're, we're ready to do that." So I'm gonna be happy on the other side if it helps with their procurement and everything. That's great. Our, director of content here calls it the hot dog problem. You have absolutely no idea what goes into a hot dog, and so- And you don't wanna know and you don't wanna know.
In this case, you do wanna know. You wanna know your data- You, you kinda have to in this case, right? Yeah. Yeah. But he's like, so often these folks get stuck with a hot dog, and they have no idea what's actually making up that hot dog. And you wanna go and take a bite, but you don't know the risk level of it. Exactly.
E- exactly. And, and I would throw a monkey wrench into that question, too. Yeah. Is a hot dog considered a sandwich? You know, that's- That's an existential debate we don't have, we don't have time for that today.
We don't have time for that. I hope you get a whole lot of Q&A in on that one, and a lot of people aiming in on that. 'Cause I'd love to know what people think. Yeah. Are you team sandwich or team not sandwich? I'm gonna argue yes, it's sandwich, because there's bread and meat involved, and that's
close enough for me. I know somebody will argue on the other side it's not two pieces of bread, but- this is where pedantic arguments- E- are fun to have. Exactly. All right. Not to turn, this conversation away from grilling, but, uh- let's, let's peer a little bit into the future here, which is almost impossible considering the reality in which
we're currently living. Yeah. But if these cont- conditions continue into 2027- Yeah what should agencies start planning differently right now to keep this continuous vulnerability in mind? Volatility. Y- yeah, the volatility. Yeah.
I, I, I think, and we, we've said this a couple of times. I, I, I think the agencies that come out of this well will be the ones that treated this as, as forcing function. It actually is, which is to make structural changes that sh- they should've been making anyway, right? Consumption-based infrastructure models have
to be a part of your portfolio. And the bigger shift, and, and the one that I think has longer legs, is recognizing there's a genuine evolution happening right now in how agencies need to relate to their data, right? And for a long time, the conversation has just been about storage management. "Here are some crates to put your stuff in, right?
We'll sell you more crates." And this kind of new conversation about understanding your data over the crates that are in it, it's a different question, right? And, you know, what is the data worth to your mission? What does it cost per workload? A- and what decisions does it enable or block?
One of the things I've said that I think really holds true is that AI has done something very interesting for data that we are recognizing now, which is we're finding other avenues of value out of the data that we have, right? The financial data isn't just beneficial to the CFO anymore. It actually can be put into a model, and put into manufacturing, and, and all of those
things, right? And then that value, that extra value shines the light on, you might not be managing the data that you need to help drive better decisions in the best way, right? And AI revealed that. A- and, and I think the cost of ignoring data management now, it's impossible to defer, and
I think the budgets are, are, are forcing that. And I'm not talking trendy framing now. I'm saying, listen, y- whether or not you're even adopting AI, you need to know your data, because you're being forced to in deciding what you need to refresh, right? And I think the agencies that are building that posture now with visibility, governance
discipline, and workload-level understanding of what their data is doing, they're gonna have a structural advantage when the next disruption arrives. Because I've said this too, which is if you need to do this stuff for AI, right, data management, data governance, data visibility, are you preparing for what's after AI? Because we've, we've gone through these trends before.
Remember when it was data analytics? Remember when it was cloud? Remember when it was virtualization? AI is just another step. There will be something after AI that will also push you into the realm of needing to understand data management and visibility, quite honestly.
And the question is whether you're positioned to absorb it or scramble through it. And I think changing your mindset on procurement is a very, very big part of that. I agree. All right. You summed that up incredibly well, but I'm gonna ask you- Yeah one more time here. What is a key takeaway?
So we, we've kinda run the gamut on today's session. What do you want attendees to walk away with kinda thinking about knowing, planning for ongoing change? Yeah. Yeah. What nugget do you want them to walk away with? Yeah, that's a very, very good question, right?
Let's summarize it all into a magical silver bullet- Exactly we can all take back to our procurement officers. That, that'd be great. Uh- So takeaway I believe is this: every infrastructure decision you make in the next 18 months is also a hidden long-term wager on how ready your agency will be for AI-driven mission requirements, right?
The agencies that treat this as a cost crisis, they're gonna optimize for the cost problem and emerge with the same underlying infrastructure posture that they had before all of this started, right? The agencies that treat this evolution or inflection point as a planning posture problem, they're gonna use this moment to build visibility, the flexibility, and the data
governance discipline that will be good for them when AI readiness arrives. A- and that's the thing, both paths are possible. The difference is just how you frame the question. And obviously, the latter one is the one that I would fall on right now, is completely shift your mindset.
If the industry is disrupting you, you have to disrupt yourself. The game has changed. You have to change how you play the game. And, the good news is, is I've, I've found agencies we've talked to with this, they are open to it. So I don't want anybody to th- sit here and
think that, "Oh, no government agen- " Yeah. No, they are actually. It's very wild to see because I've been selling federal for 25 years, right? I- Don't let my gray hair fool you. I'm old. And what I've seen, though, is that five-year
CapEx thing, because of this space, procurement people are like, "You know, maybe the subscription idea is good. We can, we can make this happen." And it's a change that needs to happen. So change yourself because the market is changing you. That would be my advice.
Wonderful. Thanks so much, Don. I really appreciate- Yeah you sharing your insights with us today. Even on hot dogs. That's good. Exactly. Exactly. All right, just a few quick reminders before everyone logs off today.
If you haven't already, check out the resources that are available for you to download, and you don't wanna miss out on a single learning opportunity. We'll also be sending you a link to the on-demand version of this training tomorrow, so check your inboxes for that. But on behalf of GovLoop and Everpure, I hope you enjoyed today's session.
And as always, we really appreciate and value the work that you do, so thanks for taking the time out of your busy schedules to join us this afternoon. Have a great rest of your day, and we'll see you at our next training. Bye. Thank you.
Planning for change in government IT requires more than just reacting to price hikes. It requires structural change. This session examines the evolution of procurement and the vital role of data intelligence. Whether you are dealing with AI implementation or legacy system maintenance, get the framework you need to stay agile, secure, and mission-ready.
We Also Recommend...
Personalize for Me
My Updates
No updates at this time.