59:25 Webinar
Ch-Ch-Changes: Navigating the Tsunami Reshaping the Data Industry
For August 2026, host Andrew Miller invites Andy Yun to Coffee Break.
This webinar first aired on August 18, 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 and welcome to this month's Coffee Break. Ch- ch- ch- changes. Ah, I did it. I didn't stutter. Or I did stutter. Or both. You're supposed to when you say that. Navigating the tsunami reshaping the data industry.
My name is Andrew Miller, your host, for the fifth year of the Everpure Coffee Break series, lead principal technologist here at Everpure. I'm joined by the one and only Andy Youn. Andy, thanks so much for being here. Joining me from Boston, but you know, it's all virtual, so that actually wasn't too hard to
pull off. I appreciate the time and the preparation. So diving in, come back a little bit to Andy's background in just a second. So as always, this is a series, the Coffee Break series. You can find all the previous ones, both at this link, bit.ly, bit.ly. Actually, I'll probably change that very soon to Everpure Coffee Break recordings.
You can see all the previous episodes. We found in general, they've aged better than you might expect through the f- the solution and landscape focus from experts in, in the fields. Of course, you can always find them and register for new ones at everpuredata.com/events.
Next month, we'll be joined by Michael Sass, lead principal technology strategist. Let me get it right. Super smart guy, super capable. We're gonna talk about cooking with Enterprise Data Cloud, making it real in your data center. It's actually a follow-up from last September's episode. We had JD Wallace on.
We were talking about some of the origin of the Enterprise Data Cloud. Now, a year later, we're actually into some of more making it real. The architecture has deepened, how we're doing products and capabilities around it, and how it continues to respond to the industry landscape, 'cause this is not a static thing, but it is more than a marketing message, more than an architecture.
Yes and yes, and also it's actual capabilities and outcomes that we help customers, help customers within your data center. If you haven't joined already, generally folks that we have on here are Everpure customers, but not always. There's a good number of folks on here that are maybe future customers, shall we say. But if you are an Everpure customer, please make sure to join the Everpure customer
community where you can chat and hang out and ask questions. And if you're only here for the drawing, that's cool. I get it. There will be a coffee lovers set at the end. The white version of this that will not have the vintage logo on it. It'll actually have the new Everpure logo on it kind of thing, as well
as a cool charging pad. As always, my name is Andrew Miller, your host. We're talking today around data and database trends, and so I think of, when Andy and you and I were prepping, it was like, I was never a DBA. The real DBAs when I was in IT operations did Informix and DB2. But hey, I needed to stand up some LAMP applications, so I figured out how to do V-
MySQL, MySQL and PHPMyAdmin. You know, and I could at least stand stuff up and blow it up and learn that select*.* is not a good thing. You know, you're like, "Oh, that's taking a really long time. Eh, what's up with that?" your background goes a lot deeper though, Andy, so please, if you don't mind, I'd love for you to introduce
yourself to everybody. Sure, nice to meet everyone. So, you know, I've spent my career as a hardcore SQL Server guy. Yes, those, plates above my head are my old Illinois plates 'cause I used to be based out of the Chicago area, and yes, I even have SQL Server plates out on, my vehicle outside here in Massachusetts. So I'm a little bit of a SQL Server nerd, you
might say. But yeah, I've been doing it for a heck of a long time, both as a DBA, DB developer. I always like to say I've gotten a closet full of those, T-shirts, been there, done that. So I've also had the pleasure of, being able to speak at conferences for a number of years now. It's well over 10 years now, which kind of
blows my mind. It's so much fun. I just love getting out there and meeting people and just sharing all the cool stuff that I've learned over the years, you know? But aside from that, I'm a hardcore foodie, whether it be high-end dining or just even hole-in-the-wall restaurants.
I love anything and everything I can get my hands on. I've, I'm a huge, huge, huge dog lover as well, and you'll get to see my dog in a little bit. But I won't get too far ahead of myself. And yeah, I'm a bit of an aspiring guitar player. My, amazing, wife Deborah, got me into guitar.
She's been doing it for a number of years, and during the COVID era, it was one of those, "Hey, I'll give it a shot," and I kind of like it. So I got my Yamaha right behind me, and, you know, got another, guitar, on, off-screen o- right over here to the left. So that looks more classical than electric, or a mix of both, or is there- It's just a regular old acoustic guitar.
Yeah. Cool. Mm-hmm. Very neat. Yeah. Very neat. Especially the foodie, I was thinking, I was actually, recently was gonna go out for my birthday with my wife just for a low-key lunch, and there's a, there's literally a hole-in-the-wall burger place that is often rested as the best burgers in town because, you know, that kind of thing.
And there's just one of them. But it's actually called, ironically, Windy City Burgers, which- Oh was this wonderful serendipity to you talking about Chicago too. It didn't even click until- literally right now. So yeah. Okay.
So let's dive in. That's a little bit of the housekeeping, but also, you know, Andy's a, Andy's a human being with a background and really appreciate. He's presented multiple times at our SC boot camp internally, and there's a decent chance if you're on this call that you've actually worked with him, 'cause he's been here for a little while and talking with a lot of customers about Everpure value.
You know, what a surprise. That's why he's on today. So diving in, as always, a little bit of a standard feel from an agenda standpoint. Four parts. First half more about, landsca- Ever- the land- landscape and maybe his background. In this case, we're gonna do a little bit of, some career lessons, from tuning queries to
building bridges and delivering outcomes, as eventually we realize that's what matters. Like, we can still enjoy the tech and not lose our technical soul, but we need to align where we prioritize our time to what brings outcomes, period. Regardless if you're a customer or a partner, you're a vendor, either way. Unexpected adventure. Next, think about a little bit of the
landscape out there, and especially, what it is for DBAs. So this is being, full disclosure for everyone joining, this is probably focused a little bit more on databases and DBAs, but I think this should apply for everything we're gonna talk about, whether or not you are a DBA or you work with DBAs, which hopefully encompasses everybody, right? So the dilemma, the paradox of, man, AI is
moving everything faster. It's a tsunami, but we also can't break the stuff that runs the business, to now flipping to a little bit of, you know, what is your top, your top five, top 10, easy to remember take on top three Everpure and databases and the value there. And then as we often do, not always, but often, hey, there have been some changes since, about
a year ago when we had Anthony Nocentino on. So just looking a little bit of the new, the new and the cool stuff kind of thing. I think with the And as always, please, we'll have time for Q&A at the end, but feel free to toss into the chat. I'm watching here, so I see hello from KCMO, hello from Denver, hello from Texas, hello from Boston.
Ha, that- that's you, Andy, so you know. A double hello from Boston. But please feel free to put questions or thoughts into the chat or into the Q&A. We'll be watching those as we go along, and specific time at the end as well. So with that, Tatiana, if you don't mind launching poll number one, that would be great.
Thank you. So the first poll, we- we always look at these. Candidly, they're meant to be a little bit of a, you know, just be real, audience engagement. So, you know, it kind of breaks up the flow a little bit. But also curious to see who's with us. As you can see, all about you, what's your role.
And then even what are some, when you have new IT initiatives, I put dumped on your team, 'cause sometimes depending on perspective, it's like, "Yes, I finally wanted this initiative to happen, and I finally got my way." Or it can be like, "I never heard about this," and the CIO read, oh, an airplane magazine. That's, like, a super old reference, but we all still get it I think. Maybe.
You know, kind of thing. So, and as this poll finishes up, we will share this back, too, 'cause the goal is A, it's helpful for us, but also hopefully helpful for you, to see from your, from your peers and- and what they're thinking and seeing out there. So leave that up in the background. That's the wonderful thing about using Zoom webinar is we can do that.
So in a little bit of, a little bit of your history, Andy, we were thinking back to, actually w- the first time that I had Anthony Nocentino on, who's one of your peers, we, we really embraced that theme, the, the odd couple. I didn't come up with that actually 'cause I was like, "I would do that to a guest," but he was like, "Yeah, yeah, this is fun." So you know why a DBA joined a peer storage company,
joined a storage company. But I think from a, from a standpoint of your background, Andy, there's a little bit of just, maybe start us off with how you got into databases, even just, like, early on career-wise, and then kinda what that led to in the first phase of your career. Sure. Absolutely. So the first thing about me,
though, is that, you know, and, you know, I just kinda wanna set the, stage is that my spirit animal is a, that of a dog. And if you wanna click, Yes. Yeah. So this is my dog Sebastian. He is a Chihuahua mix.
He is generally a very, very, very sweet dog, but 3% of him is very angry Chihuahua as well. And in many regards, that really kinda characterizes me as well. I try and be friendly. I'm extremely loyal, and all the other different characteristics and positive traits that many of us love about dogs. Wonderful. And, you know, and the thing is, dogs also
have roles. Many of them are trained for certain things, right? But for me, I didn't kinda start out that way. I started out as a generalist. My very first job out of school, was working for a small little consulting firm where I
wore many hats. I was a cis admin, I was a junior junior DBA, I was an app developer and a web developer, so I got to do a whole lot of different things. But one thing that was really cool, and if you wanna hit the next one, please, is that after that job, I got the opportunity to specialize. And, you know, through contacts actually, a gentleman that, I got to know via a
Counter-Strike server of all things, right? If you, some of you who are gamers out there might recognize that reference. I used to run a Counter-Strike server. But I met this guy, and he helped me get my first job as a junior DBA. He's like, "Andy, you know, we've been looking around for someone for quite a while.
I know your background. I know you know your stuff. We can teach you and train you, and this is your opportunity to specialize." And that's kinda where I got into, you know, hardcore SQL Server as a junior DBA. I started, you know, in the trenches doing operational type stuff.
But the interesting thing as I continued to evolve in my, progress is that I got really good at troubleshooting and performance tuning. And that's kind of the niche that I kinda discovered about myself. And I like to think that I became very good at that because of this next slide, if you will, Andrew.
I think there we go. Yeah. Yep. Is that I have a very curious aspect about me. I'm very, very curious about things. I hope you guys are enjoying this little, this little animation. This is actually me playing around with AI just this past weekend.
I had heard about some cool multimodal, n- models and such that you can run on, your own local laptop. So I took a picture of Sebastian, and I gave it some instructions, a relatively simple prompt. And, you know, 15, about 15, 20 minutes later, I had generated this. That's- Of course, I had to go through a couple of iterations, but, yeah.
And replay it again. It's so funny. Yeah. I'm just gonna keep playing while you talk 'cause I'm enjoying you- but I'm enjoying this. Well- Yes both at the same time. Not one more than the other. Yeah. But going back to the career thing- Yeah be curious.
That's actually what I think really helped me become a better troubleshooter and performance tuner, because I've always had a curious nature about me. Also if some of you all are, Ted Lasso fans, I was a little late to the Ted Lasso party, but, you know, that was, of course, a theme early on in one of the season one episodes. There's a really cool, scene with that, and I'm not gonna divulge anything else about that
if you haven't watched it. But that's something that I've really embraced over the course of my career such that when I'm tr- performance tuning and troubleshooting, it's why is this happening and really trying to dig in even deeper. And that, I think, is something that has really helped me along over the course of the entirety of my career, both as a DBA and as a developer, you know, and then going back to an
operational DBA role. But then an interesting thing happened. I had a twist or a, a left turn at Albuquerque in my career. I've been a guy in the trenches for all this time, and then I got a call from a, a buddy of mine, Scott, who said, "Hey, Andy, do you wanna become a sales engineer?" What's that?
I never knew such a thing ever existed. And, as I learned- Engineer? Sales? These are not the same things, right? Right. Eh? Exactly. So, but it, it- You know, it was one of those opportunities because he's
like, "Andy, I know you. You have the troubleshooting skills, you have the skills to dig in deep, and you are curious. And because of that, we can teach you the rest of it." Mm-hmm. I'm like, "All right, let's see what happens." And that opened up a completely new world for me where it's a really cool career path that I'm in now where it's a little bit of
consulting, a little bit of helping, troubleshooting, finding solutions, and really digging in. So I've really been able to embrace the be curious aspect of it and continue it, you know, through the course of my career. So I, I did that at a company called CenturyOne. Some of you, database folks may have heard of that company, who of course got absorbed by a
different company who I shall not name, but you might remember them making headlines, with a certain, hack involving the government and whatnot, and not so good stuff. Or involving everybody in the world, more or less. Right. Unfortunately, yes. So afterwards- I got the opportunity to thankfully come here, and I've been here for
the last, you know, five years or so, pretty much doing something very similar as a field solution architect. There's also, there's one thing when you and I were chatting a- about this ahead of time, is that the idea of kind of even expert power- Mm-hmm where just like you get good at what you do. You're not trying to be all pretentious or proud about it, but like you're
comfortable in abilities. And then eventually you just follow that into the, like what do you want to be doing in five years? I'm, I, I don't know. Like, I want to be doing interesting, worthwhile stuff and helping people. And you follow that and keep showing up and doing your role well, and opportunities turn
up if you do it well. Or even sometimes if you put yourself in an industry that's growing, as tech classically has historically, kind of thing. Yeah, very much so. And that's the funny thing is I, I always struggle with that. What, what are you gonna be doing next in, you
know, or what do you want to be doing in five years? And I'm like, I don't know. All I do know is that I want to be doing cool stuff with tech or, in this case, evangelizing, well- Mm-hmm the, the technology that I really, really love, and helping people, you know, maximize their potential with it.
That. Very much. And for those of you who are joining in, you may be like, "Hmm, what is this opening section?" Well, often when I have new guests on I like to explore a little bit of their career, not from the standpoint of like, hey, Andy loves telling his story, I love telling my story, but from the standpoint of like what are some career lessons along the way and
things that might be helpful for everyone listening. And, and of course it's more fun to start with a story than like, "Hey, here's all our stuff," kind of thing. So I think there's one more piece here though that, as you were evolving from generalist to specialized to even sales engineer and learning some of the, kind of the hard skills and the soft skills there, but,
but there is even one more key lesson I think we wanted to share with folks. Yeah, absolutely. One of the things I've found in the course of my career is that as we become very, very knowledgeable subject matter experts in whatever it is that we do, whether it be networking, whether it be storage or, in my case, SQL Server, we hear a question about
something, "Oh, I know the answer," and you want to dive in immediately and just give an answer all the way 'cause this is your subject matter expertise, right? You want to put on that cape and be that superhero. I'm so excited about it. I love the topic. Yeah. Let me go. Exactly. But there are times where, actually,
hold on for a minute. Take a step back. Ask a couple of more questions, and then listen more rather than try and dive in immediately. It's one of those funny things. It's almost like a paradox that we are almost too knowledgeable for our own good, and
sometimes we need to m- put that in check. Eventually you can of course put that cape on, but sometimes don't be overeager. Listen and slow down b- to get a better sense of the bigger picture. Well, and, and what I love, I'm, and I'm just gonna have a little bit of fun with this, is that if anything, usually chihuahuas are not known for sitting back and waiting.
So there's a little bit of your spirit animal, but you've- you've morphed into some chihuahua, some other mix in there too, you know? Kind of the, the, it's a little more calm and like, "Ah, let's sit back," so. Well- I will say, though, he is very eagerly curious about stuff. When we're walking, he's nose to the ground sniffing all over the place.
Hence the- There you go hence that aspect of it. But he is very overeager, that's for sure. The, the other thing I thought about here, and this is just, this pointing, I mean, hope- hopefully into kind of career lessons that are helpful for people listening, is I often go back to the I- the why-what-how structure- Yes in the sense of if I'm talking with someone,
and, and to me this could be if you're talking with someone and you're a partner SE or maybe you're a partner account exec, or even you're a customer and you're an admin or you're the manager, et cetera. And thinking about, okay, I'm talking with someone about this project, and are we talking about the how we're going to do it? That, that's fine. But that, we have to know that.
But if we don't have agreement yet or I'm not clear on the what we're doing versus the how we're doing it, there could be some bumps in a month or two because I might be doing all this how stuff, and then the what, oh, they thought something else than I did. Or it's maybe not even f- nailed down, much less the why are we doing it, which just goes to the Simon Sinek start with why.
Sometimes people don't, you know, it's worth listening if you haven't gone and listened to that talk, it's about 20, 30 minutes, or read the book. Both. Do both of them. But if we start with, if we're in the how layer and we don't know what the what is, if we're in the what layer and we don't wh- know what the why is, that can actually have challenges later on and actually lead to, like
for instance, when I was in, customer, I had two times that I wrote up a pretty detailed with a lot of qu- with a lot of quotes from vendors, an architecture disaster recovery plans, and guess how many times they both got funded? Zero. 'Cause I, like looking back, like I understood the how, I understood some of the what, but not enough of the why to go write the right
justifications to get it to happen, so. So anyway, back to you, Andy, for, for closing thoughts here before we keep going. Yeah, yeah. I, I, 'cause the thing that I, the parallel that I always think about, 'cause this is something I used to do a whole heck of a lot as a performance guy, is that I would just be troubleshooting minute things.
I was thinking very tactically. I gotta fix this bug, I gotta fix this query, I gotta fix this, I gotta fix this. Rather than taking a step back and th- and having a more strategic look at the workload. That's one of the things that we sometimes talk about in the SQL Server world. Are you perf tuning individual queries or are you taking a step back and looking at the
workload as a whole, trying to understand what's going on in the bigger picture, right? Mm-hmm. So it's shifting of that mindset. I grew up being very tactical, but in the course of my career I've learned how to become more strategic and the value of that, and finding the right balance. That's a, that's the other key too, 'cause you can't forget the tactical.
It's gotta be done, but where's that balance and striking that right balance? This is one of those stereotypes like, oh, that guy has his head in the clouds, or lady, either way. It could be It's not a gender thing, but it's like, oh, you get so up, like you're up here that you forget that the execution can make or break the strategy kind of thing. Okay.
Tatiana, if you don't mind, sharing back poll number one, that'll give us just a little bit of a, a little set of, of, of the mindset of the folks joining us. Thanks for everyone here today for hanging out. So we do actually have a majority of DBAs, so I'll give a kudos then to, pure marketing 'cause we, some depending on the topic, the, the mailing lists and otherwise target a
little bit more, but also this is more general. We don't build this just for whatever, you know, whatever the specific audience is. And then, it sounds like the general answer of, when you get new initiatives, it's a somewhere between weeks to months is the most often. There's at least one person out there, in case you're wondering, who did choose, sometimes
before the heat death of the universe, AKA, I'm just here, you know, kind of thing. Eh, eh. So any comments there, Andy, before I share the next one? I just appreciate that, enough of you are still at least in the weeks and the months. I'm actually kind of glad to hear that rather than something even longer term than that.
And actually, Tatiana, if you'd like to poll number two. Man, I should've had you, had you do that, so. Yes, it, it could be. That definitely is a It means, means folks are fairly capable, or their organizations are fairly functional, not dysfunctional.
So next question: How comfortable are you translating? 'Cause we were hitting on this a little bit. Complex database subjects, and actually, Andy, I added the parenthesis 'cause I was thinking about the audience, like it may be database subjects if you're not a DBA, complex technical subjects, for non-technical execs.
I do it all the time. It takes a lot of work. It kind of makes me, nervous or, no, just, just No. No, no, no, I'd rather debug, chocolate dipped preferably. So let's go into section number two. This is now, and hopefully saw on the abstract, think we're thinking about what are the major
trends out there that we're seeing, pulling the lens back a little bit, so both the DBA's paradox of don't break it, gotta keep the lights on, versus we have to lean into new technology or else we know that we're gonna get left behind, candidly. So I think, yeah, I'll j- I'll just put them up here and, and toss it to you for each one without a long lead in, Andy.
So first one we're thinking about, replatforming/repatriation. So there's, of course, two aspects to that one. I'll do the, first one first, replatforming. Who here is on VMware? Who here is trying to get off of VMware thanks to what Broadcom has decided to do over the
last couple of years? That, of course, has been a huge trend in the industry whether we all like it or not, right? People have been scrambling for the last couple of years, just trying to figure out their next step. Some folks have been able to make that step and put, their game plan into action, whereas others are still trying to figure out their
way of what are we gonna be doing next, or maybe they're waiting on another vendor that is their first candidate, but they need certain- Mm-hmm capabilities in place that VMware already used to bring into, you know, already had, you know, available, right? So, you know, replatforming is definitely one of those things that has been a challenge. And one of the aspects of replatforming that many people are considering is the cloud.
But that's where the second one comes into play because there are folks who, have migrated to the cloud years before, and they're getting back out, hence the repatriation. We have a lot of customers who, migrated database workloads up to the cloud and have found that the cloud is good for some stuff, but is not as cost-effective for the high-end
database workloads. Can it perform? Sure. But is it gonna cost you? It's not That's where people are getting burned because I don't know about you all, but I've seen a lot of companies that did a blind lift and shift. It's n- in, to really y- you know, see cost, savings in the cloud, oftentimes you either
need to pay down technical debt, you need to fix, poorly performing queries, stuff that we used to just throw more hardware at when it was on-prem. And now that people are paying the, paying the price up in the cloud, they're ripping it all back down and, you know, repatriating it. But some of those people are, you know, who are repatriating, there's others that are
saying, "No, now we gotta move to the cloud because of Broadcom." So it's an interesting, crazy set of dynamics depending on where you and your organization is, in that process. Yes. Andrew, what have you seen? Consumption-based pricing is good until you realize there's a meter running based on an inefficiency you'd forgotten about five years ago.
Five y- yes. Yes, and the, the buzzword, AI. We all have to have AI, and frankly- Ching, done yeah, it's one of those things where, I mean, I've personally embraced it, but I do see its, you know, pitfalls here as well, but I definitely hear from many different customers where their executive's like, "We
want all in on AI," and then the rest of us down in the trenches are like, "Okay, how do we actually get there? What do you want us to do with AI? What, what are, what are the practical goals and outcomes that you want to drive us to?" Rather than, "Let's all use AI." Well, that's kinda great, and then, okay, well, if you want us to use AI, well, we might need some upgrades and infrastructure, but how do we
shoehorn into what we already have, right? Mm-hmm. And, you know, many of us have Like, I live in a, a, I live in a house that's over 100 years old here in Boston, right? And, you know, some of the walls have been redone. It's, you know, it's been renovated and remodeled, but fundamentally the one thing that's stayed the same is that foundation.
Mm-hmm. And of course me as a database guy, what's that same thing that oftentimes stays pretty much the same over the decades? Your database. So how do you shoehorn AI capabilities into that, especially with, you know, vector capabilities coming to the table that we now need to, bring into play in order to leverage AI?
How do we embrace all of that? And I don't know about you all, but I've heard from plenty of people, myself to an extent as well, it's kind of overwhelming. It's, it's interesting how each of the I'm not gonna say traditional, 'cause that's not meant to be There's, like, no negative in there.
Just, like, the longstanding databases out there. Let's just pick on SQL Server and Oracle and Postgres. Mm-hmm. Everybody's doing vector in some way. Yes. But even before that, actually this is now probably a year or two back with Anthony, we're talking about bringing object in.
We were talking about more stuff with external tables, you know, kind of thing, so you can actually see the data where it lives and not have to bring it all in. But then that even inter- and I, now I'm pulling ahead of one of the themes later on. Oh, well. You know, these, these do blend together. But then you even get into the sense of, okay, as I start to think about this, if I'm
referencing from my database external things, cool, I didn't have to move all the data, 'cause data gravity's painful. But I now have a whole different re- dependency chain question from a operational stability question, too. It's not just my database is this homogenous, good, in a good way, thing from a stability standpoint.
I've got to think about the multipliers out to all the other systems. Which maybe brings a little bit to data sovereignty and localization, but yeah, yeah. Well, keep going. Yeah. No, what I'd say to that is this, is that, you know, depending on where you are at in the world and/or what industry or vertical your organization may be in, there's a lot of new
laws now where we have to deal with, you know, data security. Or for example, like, I know of many European companies that are having to get out of the cloud because of different regulations that they're now subject to about where their data may reside. So if it's up in the cloud, they have no control over it, for example.
Hence the data localization, example, for example. And you know, a lot of these different things, you know, all fall into feeding into the, the umbrella I like to think of as data governance. What are we doing with our data, and do we have visibility around our data? It's a big challenge that a lot of folks, you know, kind of realize that they have, but they
really have no idea of how to really navigate this particular challenge. I'm trying to think, what is it? You know, it also, especially if you're in, you know, you know, pub sec, if you're in, you know, SLED or FED, for example, this is something that really impacts a lot of organizations.
Andrew, what have you seen around this? Localization can also sometimes just mean fragmentation, where especially- if you're a multinational company, now we have requirements to have data in certain regions, not for the wrong reasons. Like, because if there's regulation, we need to comply with it, but it's not for inherent
technical architecture reasons, and then those can drive technical architectures that may candidly be, be less ef- le- may be less efficient. There's even a, a fun pull forward here. We're gonna talk as we go along the way- about Everpure Data Intelligence, and that actually started out of the startup that was 1touch, actually started out of this trend five years
ago, or whenever GDPR was passed into law or talked about in Europe. So this is, to a degree, a trend that's been out there, but we're seeing it accelerate. I think I may just build up the last two, Andy, and let you, let you do both of these together, and then I'll, and then we'll finish up- Sure the section. Please. Well, bringing compute to data, we already
kind of touched upon that, but it's- Yeah. My fault thinking about the Oh, no, it's- I couldn't help myself it's fine, 'cause we're gonna be doing that a lot this, this next hour, trust me. But it's So bringing compute to data. Many of us grew up with having to move data around for different purposes.
ETL, right? You know, so it's like I gotta extract it, I gotta take it out of my operational database, move it over here to my data warehouse for, to, do analytical workloads, and maybe I gotta move it over here to a reporting environment. The trend has really reversed.
We're trying our best to keep the data in this, just a single place rather than trying to make copies of it and move it all around. And rather have the compute be able to jump over and just access the data wherever it happens to reside. Data lakes, data, delta lakes, all those different terms- Lake houses and whatnot that
we hear. Lake houses, yeah. All those different terms that we hear all really encompass this, this overall data trend. You know, we, we talked about, mentioned data virtualization and, and external tables. That one is specific to SQL Server. Or what I mean, there's external, forms of external tables in other elements.
But it's the idea that, hey, within, from this perspective, I can now query stuff all over the place rather than having to extract and transform it and then load into my local database. So that's one of the big trends that's also been, you know, you know, impacting the data landscape over the last couple of years and will continue to do so.
I don't see this changing at all. And then changes in buying decisions. This kind of leans into also, like, data gravity 'cause the, the So data gravity is the more data that you amass, the more people will be drawn to it because then you can do more interesting things with it.
And more people oftentimes being people further up the food chain asking more questions and being more curious about what they can do with their, you know, with their business and the business direction that they're going in. And they want more answers, hence why they also want things like real-time AI in their databases so that they can go digging for more insights even faster.
All of these different things tie in together. So I'm actually gonna do a little bit of a callback to if we go back to January of 2026, Ian Saunders- we were covering AI. Yeah, on Analytics Insights. Sounds fancy. Curing amnesia, sports cars and semi trucks.
One of the things there that we hit on as well, a comment that, actually a talking point that, NVIDIA, this guy named Jensen Huang, you might have heard of him, and other people too have made about that most businesses have all the data they need to be successful from an AI standpoint. It's more about can it be transformed?
That could be a word, fancy word for vectorized, so you know all of them can access it. Can you actually weed out the good data from what's, from, from what, what's authoritative versus what's old kind of thing? So this sense of, like, that there's enough data there, and if anything This is what I was
thinking as you were hitting on this, so a little bit of commentary on both of these two, Andy, if that's all right, is that bringing compute to the data, we've thought about this for a while, but as data explodes even faster than it's been growing, I know that's a cliche, but that ramp only in- keeps increasing- We have to do that, that much more and more, 'cause we can't actually even pull off moving all the data or copying it to wherever we want.
But it's like the only way we can cope with this at all is to handle the da- data where it's been manifested or generated kind of thing. And then buying decisions, this one to me is fascinating because what thinking about And, like, I think about this with I, I report to a gentleman who's a, who's a VP at Pure, okay? So, but a s- a super good guy, too.
Phil Zacaro, I'll give him a shout-out. I should always do that for my boss, right? But the sense of when I'm working with him or with other folks in executive leadership, I try and think about, like, what is above and below their cut line of what they think about or what is worthwhile for them to be thinking about so I'm a good, you know, employee and
I'm bringing the things to them they need to know about or bring solutions or heads-up, I'm doing this thing. But that cut line is shifted or moved as it relates to candidly infrastructure, because a lot of the things that were seen as commodity and non-differentiated five years ago, either maybe now is they're just actually really hard to get due to supply chain things, or they're differentiated enough that they might be the
make or break in can your AI project go from p- POC or pilot to production without falling over. So, interesting trends. Thank you, Andy. I appreciate you walking through this, and I, I know you had good response when, you walked through this.
Oh, it was a p- with a partner I think a couple weeks back. It was Yes. Yeah. It was good. It was a, a fun little local event here in Boston a couple weeks back. Yep. Awesome. Tatiana, if you don't mind sharing back poll number two, and, just to let everybody see it for a minute.
Thank you. So th- there's nobody here who's wholly uncomfortable. Good for you. That's great to see, you know, kind of thing. But I think it is, we're just kind of split between, the ma- we actually have a m- a, a majority on somewhat comfortable, and then, you know, kind of split between the other ones.
So, good to see, 'cause this, these are, these are critical sell skills without losing our technical souls and even the enjoyment of what we do. Tatiana, if you could launch poll number three. Thank you. And of those five trends, we're curious to hear from you all, which of these is relevant to your company.
And man, we went back and forth. Do we make that a multi-select or a single choice? We decided, eh, we'll make it a single choice and just so, you know, so StackRank can see. And then also how much urgency, pressure are you under leadership to integrate AI into your database stack? Here, got a little bit
more specific about that. 'cause in some cases it might just be, eh, the AI folks, they're off over on the side doing something. They'll come back and talk to us when they do whatever they do. Which might work. It helps with rapid incubation, but it may not help with going to production, so it's always
a, always a trade-off. Section three. So help me remember, Andy. You've been here four years? Just Three? Five. Almost five. Five. My bad. Close to five, sure. Oh, shit. Got it.
Okay. I, obviously my LinkedIn stalking is not up to, up to snuff. I should be better at that. But so we were thinking about this. So as you've been here, a good bit of what you do is going deep, but at the same time you don't start with, "Let me hit on this random, deep, amazing thing about what Everpure does."
You usually start at a little bit of a higher level. So if you don't mind, maybe this is kind of like the, Let's go back to, this doesn't apply now, college, and I was gonna say CliffsNotes, but that's been redone by just ask your LLM of choice, "Give me the two-page version of this book." Okay, fine. But if you were doing that version for folks about, the value for Everpure in databases, I
know you usually like to group it into three categories. Yeah. So if you don't mind, kind of take us through the top level categories, and then I'll, I'll see if I can build out the slide and follow you along. We'll see how it- Yeah, absolutely. So the thing is for me is that of- especially when we're talking about those complex topics,
that poll question that we just had up there a little while ago, I like to try and distill things to something much more higher level, and you wanna have that list STaaS short and as simple as possible. 'Cause the thing is, frankly, most of us will forget half to two-thirds, if not a higher percentage, of what we hear. This is something that I, came to learn and understand as a conference speaker.
So for me, what I always try and distill things down to is, are these three value points or principles that, you know, we bring capabilities and tools to the table where we can help you accelerate, simplify, and reduce risk. And all the different things that we offer can fall under one or more of those different categories is what I like to argue.
So accelerate, simplify, reduce risk. Four easy words, three principles. So acceleration, you know, u- under that the easy one of course is, you know, sub-millisecond latency at any scale. We are really fast.
We know that. That's table stakes at this point, right? That's the easy one. But let's think about what are we accelerating? In the context of a database workload, that's accelerating a front-end application that are used by your customers. I don't know about you all, but when I have my smartphone out and I'm doing stuff, and if
something suddenly, you know, pauses and takes five, 10 seconds, I start getting annoyed. I get impatient. 'Cause frankly, a lot of us, myself included, our, our, our patience level is a fraction of what it used to be, right? Mm-hmm. We need to have stuff well now. That's what we've come to expect.
And if your underlying infrastructure can't do that, that means unhappy customers. And that also leads into the second element, too, of the predictable, consistent, latency and such- and all of just the other capabilities that we bring to the table that can help accelerate other workflows. The last one, the zero-touch snapshots.
You know, oftentimes even internally you have your I- it's not your outward customers, but, you know, maybe if you're an infrastructure person, you're supporting developers and QA and other teams, reporting teams, right? These are pseudo customers of yours. So to be able to take data and snapshot it, present it to other environments, that's
accelerating internal capabilities and internal workflow. Simplify. Hold on for a sec. Like, something has happened with my- It was a little bit odd with your audio. Hopefully you're on a weird prompt with, my, Zoom here about, my mic switching up. But as long as, everything sounds okay to you, we'll keep going.
You are now good. There was a blur- a little bit of a blip before, but you sound great, so keep going to simplify. Okay. Yeah, that's really bizarre. Anyway, I apologize. It's live. What do you know? So, simplify. So that's the thing. So the thing about Pure is this: I'm a
database guy. I've never been a storage admin my entire life. I can run a Pure//E environment, right? So if we're able to do things in a much simpler way, such that you don't need a 40-hour training class and, you know, a year or two worth of experience and a team of 10 people in order to manage a storage environment, isn't that a powerful thing?
Or can we help simplify existing workflows because you can now do things in a programmatic way, rather than introducing human risk if you have to click through a GUI and make sure that you click this, click this, click this, checkbox that. Oh, y- 'cause how many times have you missed a checkbox on accident even though you may be following a set of instructions like a run book in a disaster recovery scenario, where
you're freaking out because you lost your primary data center and you're trying to fail over, right? So wouldn't it be nice just to have all that pre-scripted and automated? Again, an example of simplification, but also an example of risk reduction and acceleration, because if I can fail over even faster in a DR scenario to get my business up and running
again in a DR scenario, that's a huge benefit. Again, a- all examples of how all of these different value props, kind of, relate with one another. You wanna hit the last one? Yeah. One thing I love here, if you don't mind, is that there, obviously we, we group these for, to help f- track them, but depending on the lens that we look at it, anything in
simplification can actually be reduced risk, and often reduced risk is simplify. Like, the best steps is the one that I don't even have to do anymore. I can't mess that up, hopefully. Yeah, yeah. And that's, that's actually a perfect lead-in to, for example, with SafeMode mode and immutable snapshots.
If I just set up snapshots once an hour, now I have that extra layer of protection. It's a fire and forget sort of thing, right? So that way I at least know that I now have a safety net in case of, say, ransomware attacks, for example, or even just database oops. I have a great blog post out there about doing object level restore with snapshots, 'cause
that's f- something that's not natively available in SQL Server, but we can at least help facilitate that. I have a 13-minute demo video just walking through the, "Here's how I would do this even if I didn't automate it ahead of time, and if I just had to click through the GUI." So, being able to just do that is, again, risk reducing.
It's giving you additional tools in your arsenal. So- And even- Yeah, go on yeah. Oh, no, please. So, so when we're looking at this, I, I was actually thinking about there's a classic one, and I, I could see it here. Often when, when we talk about looking, working with executives or people that are at
business level, we talk about there's make money, save money, mitigate risk. Mm-hmm. And ironically, I don't know that you were inherently trying here, Andy, but if you were, you know, gold star. But I think everyone here, if you're listening, you say you accelerate, this is stuff that can often, not always, but often can actually make you more money.
It can help reports run faster. You can be more nimble about responding to the business, support new initiatives. Save money, often that's simplifying. Okay, right, you know, we can do things faster with less cost and potential mess-up. And of course, mitigate risk, well, that probably, that probably maps the most closely,
but there's even some blur together. And if you're thinking, "Hey, some of these are the classic Pure values," you're right. And actually, one of the first three use cases from a, ever, from a Pure standpoint, not Everpure, but Pure, like when we started out, one of the first three use cases that was actually targeted was databases, because of the need for performance and the reducibility
of the data. That hit really well in the early days of Flash. And if anyone is curious, you can actually Google on this and find it. It's, actually still up on Bright Talk as well as some other places. The very first year of Coffee Breaks, this is 2021, had your peer on, or previous peer and
friend, actually- Mm-hmm Argenis on, and we were hitting on some of the same themes. Now, there's newer aspects to it, and you have new stories. Good for you. You should. Even when we had Anthony on, you know, last year, similar pieces. So it's not that this is, like, Pure continues to move and improve.
That's literally section number four we're gonna do in a second. Mm-hmm. But th- there's a consistency of the things that we're talking about and continue tuning to make those richer and deeper. I wasn't trying to cut you off on anything here, Andy. Oh, no. So please bring us on to the section.
Keep going. No, we're good. Cool. No, we're good. We can move on. Okay. Tatiana, if you don't mind, let's, share back poll number three, and we are in the home stretch. We'll be roughly on time here today, so thanks for enjoying. Actually, saw a couple more folks join as we
went along the way. The recording will be a- available. But the f- five trends, this is interesting, actually. I would not have guessed that bringing compute to data would have 0%. But, data sovereignty localization, real-time AI, operational, tied, you know, number one, and then the other two. And then let's see here.
Urgency and pressure. So looks like the majority are a- being asked to explore, but not with a, "You gotta do something, anything, I don't care what, something that I can tell the board." But there are some of you out there like that, the time travel, you know, "What's the, what's the best time to plant a tree?" 20 years ago. What's the best, second-best time?
Today. You know, so that's science. That's your AI initiative, sadly. Okay, feel free to comment, Andy, if you don't mind, and Tatiana, launch poll number four, and we'll leave that up here. Okay. Hey, the polls get simpler as we go along, you know.
Of course, 'cause let's not, hopefully we're not worrying anybody out here, but, so we hit on three themes: accelerate, simplify, reduce risk. Curious which of, which of those themes resonate to you the most. And if you have some thoughts on it, please go ahead and put them into the chat, and we'll I'm actually gonna make sure that I'm looking at that so it's not in the background.
Good deal. Okay. So last section. Hey, we're almost done. If you're staying around for the drawing, hey, you're just about there too. That's cool. So what is new? There is, there are definitely new things in this space.
The database as a workload category, and I realize I haven't been ha- calling out the, you know, having some fun. Hopefully been seeing the little subtitles up here. We kind of left them up there as Easter eggs, so Wait, it can do that now? Really? It can? So the first one- I'm sure nobody,
nobody has heard of a little company called, Oracle. Oracle. But for everyone that has- which is everybody, hey, we've got some new stuff here. Andy. So one cool thing that we've now bring ta- brought to the table is what's called a Wol Data assessment.
Wol Data is a company that we've partnered with, which will take AWR reports, which are basically telemetry-type reports that give us all sorts of information about what's going on in your Oracle environment, and we will go through and process them and figure out really not only just what's going on in your estate today, but do assessments on them. And it's expert-led, where one of our field solution architects will actually help you out
and even give you some guidance, depending on what your organization's, goals happen to be, where you happen to be going. If you're looking at adopting AI capabilities in Oracle, for example, are you on the right roadmap to get there? Another of the use cases for using this, that my, colleague, Thomas Stutzman points out is
that, "Hey, are you using encryption? Do you know if you're using encryption or not?" Because, transparent data encryption in the world of Oracle is something that actually costs you extra money. Did you know that, you know, the group over here, actually turned it on without you knowing it? 'Cause stuff like that
unfortunately does happen. The right hand and the left hand are never talking to each other, right? Many of us are stuck in organizations like that. So this is a really cool offering that we happen to have that can really give you a full-blown, a, a full-blown perspective of your entire Oracle estate.
And then once you have that baseline, now can help give you that roadmap forward. And again, we can s- also share advice and guidance with how you want to grow or perhaps even shrink that Oracle footprint. This is where And, and there's even been a little bit interesting thing as far as if you think about previous Coffee Breaks, we focused often more on SQL Server.
Sometimes with our folks, our friends in EMEA, or Europe, Middle East, Africa, okay, we've started seeing more Oracle workloads. But frankly, from a landscape standpoint, there's a huge amount out there, and if anything, we're seeing increasing uptake in America, US, Canada, South America's there, around Oracle. So I wanna make sure to highlight this.
Now, probably if you tr- if you're attending Coffee Break, you pay attention to Pure a little bit, just being real, and so you probably heard that we acquired this company called OneTouch, now renamed Everpure Data Intelligence, so just to kinda contextualize that. This operates at a different level of the stack.
We have capabilities where we can do interesting things by understanding the data that lives on a Pure Storage array. But this actually plugs in at a higher level of the stack, and even at a database level, and therefore can be agnostic to whether the data lives on a Pure system or not. A little bit similar to how Portworx can provide value in a cloud-hosted Kubernetes
environment and on-prem. But Andy, I know that you were thinking about this specifically kind of in a data, database context, so please. So for me, I think this is an amazing solution for data governance. Who, you know, who out there actually knows where your PII data happens to be?
And, you know, can you actually give documentation of, "Yes, I know my PII data is in here, here, here, and here"? Because if you're attacked in ransomware and, you know, the authorities show up and start asking questions about what data is where, can you answer that? This is something that can actually help you do that and catalog all that without, as much
manual intervention. One of the things that I dealt with, at my prior job before coming to Pure is that we had a product, that, you know, basically tried to help you create a data dictionary. It was actually really cool, but the problem was this: was this: It still took a human to fill out everything.
So you still had to go clicking through the different columns and tagging them and categorizing them. It w- Once you did all that legwork, it would generate some really cool reports and HTML stuff, but it still required, worker hours to do this and, you know, subject matter experts to do this. Whereas Everpure Data Intelligence gives you
that jumpstart, that headstart to at least do an initial catalog, because of course, being AI-driven with semantics, is able to figure out a lot of it. Mm-hmm. Yes, there will absolutely still be stuff that you're gonna need to review, but hey, at least now I'm able to cover 80% of my estate and now just have this checklist of, "Here are the things I need, just need to double-check," but otherwise, it's done the grunt work for me.
This is the next step to be able to finally get that data dictionary that so many of us d- data professionals crave or wish we could have, but don't have the hours to put together. So there's even, And whenever I say this, it's like, oh, we, we could've added more slides. I'm just gonna use slides more for, for illustrations. But what I find really interesting about data intelligence, as you're referring to, look,
Andy, there's, like, multiple audiences here. You're going in from a governance standpoint, so there's a whole aspect to that. We sometimes find that this, in that is actually some of the origin of OneTouch, this startup, before we bought them and they were renamed. Or they, they chose to be acquired, I kinda think, 'cause it started out of GDPR
environments in Europe. There's also a whole aspect where there's an AI buyer for this or projects where it's like we need to find the right data to feed into our LLMs, you know, kinda thing. Otherwise, that's gonna be a challenge. And then even from there, you can even say, look, maybe there's just a general from an IT
organization standpoint, like, "Where is all my data? Where, where does everything live? I need to understand this stuff." So this is where it's this interesting thing. It's even g- being very real, actually a little bit of a challenge at times for us as an organization, because we sometimes have very different personas inside our customers
that care about this, and we have to understand how to speak the language of what that person cares about kinda thing. But it, that's a good problem to have. That's a really good problem to have, you know, kinda thing. So let's do, bringing this home for the last one, thinking about a standpoint from a
performance perspective, and a little bit of a multi kind of performance thing, if you will. And actually I'm just making sure I've got one thing. There we go. Okay. So we wanted to kinda group a couple things in here. One is, you may have seen from Accelerate, we talked about FlashArray XL 190 with Purity Turbo.
This is frankly focused on database workloads, not only, but often the most serious workloads in the data center are databases, you know, kind of thing. And the interesting thing here is that one of the things that we've done here, this does have XL 190 has a very large read cache to help with very cache-friendly workloads, 'cause nothing's faster than that kind of thing if it's a cache-friendly workload.
But then also what we did is we said, "Hmm, let's go look at a dual-controller architecture." All the things we've said for a very long time about never, not having any performance impact or downtime during software or hardware upgrades, still true. But what if we could intelligently leverage the second d- controller for certain scenarios? A little bit of a have your cake and eat it too, where you understand the application
workloads enough to fit. So to kinda illustrate that, you know, there's this great, slide from Andrew Silifont. You know, for instance, you've got an operational load, you know, however many things going on, your core stuff, you know, some backup, ETL. And then, you know, actually you've got more stuff coming in.
Like for instance, I don't know, an AI workload for instance that needs to access the existing data, 'cause that might actually happen. So what if you could intentionally have a boost for certain non-critical workloads, the stuff where if that goes away during a failover, it's not actually a problem. We actually had a little more fun with this.
If you ask Gemini to, "Hey, go make this image cooler," you get something like this. Can't help an AI joke. So, so you did, the, maybe the better version, Andy, with actually making your dog even more cute. And I did the, I don't know, this is like hot ice or something or whatever. I don't know. But, you know.
Any other thoughts here before the, for the, before the last one? I know you had one thing I think you wanted to highlight. I was just r- remembering, what is it? Wow, I suddenly blanked on the one thing I wanted to highlight. Wow. We'll keep going. Sorry. We'll come back.
Let's just keep going. A twofer in the, performance side. One is what we're doing with the XL 190. It's not database specific, but it definitely relates to databases. But then there's also sometimes where, it's possible that I might have heard a DBA at some point in my career say, "You can pry my backup control out of my cold, dead hands," 'cause-
And I may be one of those 'cause they may have been the person on the hook from- Yes their boss or the executive to get stuff back online. But we've got something cool, cool there too. You remember I mentioned earlier about having that closet full of shirts, been there, done that, and yeah, I've unfortunately have been through, my fair share of
disaster recovery events. I remember one time that, both of us DBAs, we were at a conference. We were called out because the SAN admin who was doing some cleanup actually blew away a, a LUN on a production database server. We had everything pre-scripted to code out to in order to do all of the restores, but as we
discovered, we brought the SAN to its knees reading from back, from the backup SAN to the primary SAN. And we, it wound up like, 'cause it was terabytes of data, and this was about, oh, 12 years ago now, so more spinning disk era. And it took about, oh, I don't know, eight days to recover everything. It was all ready to go. Yeah.
Progress bar, progress bar, progress bar. Pretty much. We actually had to stop it and then we asked the business to reprioritize which databases do you want first. We changed up our code in order to then get those online. But they're all production databases.
But it was literally just go and then, "Okay, sorry, the SAN is down to the bottom of the lake." So, you know, this is something that I, believe me, I really wish I had because to be able to restore up to 113 terabytes in the span of an hour, I would've had that environment backup and running in less than an hour. You know, of course that was the, data scale at that point, but I still know the, company
because they're a customer now. So they would able, they still be able to recover, you know, that same, database server in about span of two or three hours. That's about how large it's grown in, those subsequent years. And these are the capabilities that we're bringing to the table.
This is just an example of just frankly raw horsepower. And sometimes that's all you need is just raw horsepower, and that's what we got here. And what I love there to highlight is the backup is very solid. 20 terabytes an hour. Mm-hmm. We're in eight instances.
You know, if you're trying to read this graph, you know, one versus eight instances. So backup is faster with instances, but the restore, it's even faster. And I'd argue today that what's, that is what matters more. It's more about how fast can you restore versus how fast can we backup. 10 years ago, more about backup and restore.
Today, more about restore or cloning environments or whatever else kind of thing. 1,000%. Okay. With that we are It's crazy. It's hard to believe. It's been another hour minus seven minutes, 53 minutes or so. Thank you so much, Andy, for doing a, an
amazing job walking us through. I, I appreciate it. Both and as those, for those of you who are listening, this is never scripted. It is live. We do spend time preparing. So Andy, thank you for the preparation time and kind of thinking through how to sort
things around in a way that hopefully is, worthwhile both to today's audience, 'cause actually we've continued to pick up people as we've been going longer, as well as to folks listening in the future. 'Cause usually you find these, these are not totally timeless, but we get a, a year or two or three or five out of them and still, and they're still relevant. I'll be curious to see how the trends come
about in, three years from now if someone watches this in 2030. And they're like, "Ooh, yeah. So like, was that Andy should've quit his job and bought some stock and st- played the stock market?" Or like, "Eh, yeah." Or me too, either way. So you're at the end, so bring it home a little bit. We'll have a little bit of time for Q&A, and a
last word from Andy. Tatiana, if you wouldn't mind launching the fourth poll results, just so we share that back for folks just to be good about letting people see it. So the mitigate risk- Interesting reduce risk was the most interesting, which is not surprising, 'cause that often comes up. That's the whole, I can't remember if we're doing cyber.
I think we're doing cyber in October. That's the reason that c- cyber resilience keeps coming backup as a topic. It relates there Next month, please do make sure to join us, with Michael Sass, Principal Technology Strategist, talking about what we're seeing in changing customer needs, as we just talk about Enterprise Data Cloud with customers, about how speed and efficiency
matters more than ever, and even maybe a little bit of the, the risk of being replaced as IT practitioners if we're seen as the department of no, but how Pure can hopefully, how Pure can hopefully help you with that. There's actually gonna be a second dog analogy coming up, next time- Yes from Michael, so hey, two months in a row. And then he's actually, he also likes, has
some fun with being a chef, so c- we're gonna talk about cooking the EDC souffle- Ah making it real with fleets, topologies, et cetera. Then how you present this, pa- the outcomes that you provide to your executives. So we'll go down, like, what are the building blocks? What, what do you provide? But if you waited around for the survey, hey,
we got ya. So I'm looking over here. Michael Yu. Try and be good about saying last names, just like earlier when I was saying my, birth- I went out for my birthday, and I was like, "I'm not gonna say what day that is on, on the wide open internet." That feels like giving away a security question.
So Michael Yu, not saying your last name, you are the proud winner, or hopefully happy winner at least, of a coffee lover's set, where you see you've got the, the newer Yeti mug, and you can charge as well. And I think I'm looking here. We're gonna give it a second here if anyone has questions, but I think I may toss one
thing at you, Andy, here, 'cause we've got a minute. Okay. And we didn't script this at all, so, you know, hey. So when, when you're out and about meeting with customers, what's the most common question that you get from folks? I'm curious. And then we may, then we may bring it home.
Interesting. I gotta think about that one, 'cause the thing is, I get all sorts of wide variety of different questions. But I guess the one that I really get more often than not is how to make more effective use of snapshots. Mm-hmm. Because frankly, a lot 'cause I do talk to
more database audiences, and a lot of us DBAs- Yeah over the years have been burned by snapshot implementations- Oh, I'm sorry. Yeah and other legacy storage implementations. Yeah. So we're very, very, leery of snapshots. You hear snapshots, and you wanna run the other way, right?
Don't try, yeah. Um- Or fool me once, shame on me. Fool me twice, shame on Oh, I got it backwards. Yeah, so yeah. Exactly. Exactly, and I too have been burned by, snapshot implementations in the past. So frankly, that's actually one of the most popular topics.
The second one that I would say is also, you know, mixing and matching HA and DR. Mm-hmm. 'cause of course we bring in a couple of different, capabilities to the table, like active DR, which I forgot to even talk about how I like to think of that as being the easy button for DR purposes, right? If I have a whole collection of, servers that, you know, maybe I'm doing, you know,
SQL Server availability groups, you know, database-specific replication. But then I have this whole set of other stuff that I'm not really doing as much with that. Well, you know what? Throw that into active DR. That's very much an easy button that can get you, impro- that can improve your resiliency posture in a tremendous way, right?
So that's one of the other conversations that I have very frequently with customers. So I'm actually, I'm really glad you brought that up- 'cause we did forget to, talk about that and why I like, calling it the easy button. I, sometimes I've said over the years, 'cause this, this for me, just being real, I started with, well, a company that starts with N and ends with P, but snapshots on that platform
was really good. But there still were some drawbacks. But so I like to talk about science snapshots on FlashArray, FlashBlade, is what snapshots wanna be when they grow up. Yeah. Like, you don't have, you don't have
dependency chains. You don't have a random waffle iron process that kicks off when you delete them. So there was good stuff, but this is, they're much more mature. And even I'm gonna borrow from you/Anthony/Graham, the multiple folks on your team, like, the idea that you can bend space and time.
Like, how long does it take to restore 100 gig database on Everpure with a snapshot? The exact same time as basically it takes to restore a five terabyte database 'cause it's just moving the metadata pointers. That's pretty amazing. And we, and we can have fun with playing with that, so. Okay.
With that, we are at the end. Andy, any final words? I don't wanna cut you off, but we got a minute, but you talked a lot, so if you're out of words, that's fine too. You know I get it, but thank you again so much. I will just say if nothing else, always remember to be curious.
I love it. So please join us again next month, with Michael Sass, myself, and on behalf of the entire Everpure Coffee Break team, 'cause there are a bunch of folks behind the scenes that help make all this work and do the, send out the gift sets and registration. And Tatiana always joins us, slides every single time to make sure that things go
smoothly, as they almost always do, as they did today. Hope you have a great month, and we'll look forward to seeing you next month. Thanks again, Andy, and everyone joining. Thank you. Have a good day. Take care, everyone.
Andrew Miller
Lead Principal Technologist, Everpure
Andy Yun
Consulting Field Solutions Architect, Everpure
For August 2026, host Andrew Miller invites Andy Yun to Coffee Break. As a DBA himself with a long career in the data + database field, Andy now talks to Everpure customers (or sometimes “future customers”) every day about data and database challenges and solutions.
We’ll wander through:
- Andy’s history with key mentoring lessons (both those he has learned and has helped others with), the value of soft skills in a deeply technical role, and why a solution architect works at Everpure.
- Database & DBA landscape - simultaneously keeping up “business as usual” while adapting to the AI tsunami that’s here whether we like it or not.
- Andy’s top DBA & database tips on Everpure from chatting with DBAs and customers every day.
- What’s new in the data landscape at Everpure - everything from an Oracle 26ai readiness assessment, to mind-blowing SQL backup throughput numbers, to how recent Everpure acquisitions impact you.
Overview of Coffee Break by Everpure. Who knew that the best coffee break conversations would happen online? Each month for the last five years, the Everpure Coffee Break series invites experts in technology and business to chat about the themes driving today’s IT agenda—much more podcast than webinar. This is no training session, either; it’s a freewheeling conversation that’s as fun as it is informative and the perfect way to break up your day. While we’ll wander into Everpure technology, our goal is to educate and entertain rather than sell.
We Also Recommend...
Personalize for Me
My Updates
No updates at this time.