34:43 Webinar
Smarter VCF Deployments: Simplifying Storage with Everpure FlashArray
Join us for this Expert-led Demos webinar and learn how Everpure FlashArray™ simplifies and accelerates VMware Cloud Foundation (VCF) deployments by serving as a high-performance NFS-backed datastore.
This webinar first aired on May 26, 2026
The first 5 minute(s) of our recorded Webinars are open; however, if you are enjoying them, we’ll ask for a little information to finish watching.
Click to View Transcript
Hello, everybody, and welcome today's expert-led demo. I hope everybody had a fantastic weekend. Matt, there's nothing like starting a Monday on a Tuesday. It's a, it's a Newsday? Yeah, right. Is that what that is? It's a Newsday.
Yeah, right. Well, welcome to Smarter VCF Deployments: Simplifying Storage with EverPure FlashArray. I am your host today, Jason Langer. I work here as a senior technical evangelist slash tech marketing, and then I'm joined by my good friend, Matt Webb.
Hey, Matt, how are you doing today? Doing good. Doing good. And I'm a field solutions architect, so what that essentially is, is a subject matter expert in virtualization in the field, visiting customers, working with customers to leverage our technologies and things like VMware, for example.
Right. Right. VMware is not dead yet, right? Despite what everything on the web says. Right, right. Well, for the audience here, just know that Matt is the smart guy. He's gonna be walking us through some VCF constructs.
And then, right, Matt, I think you're gonna actually do a demo. It's called expert-led demo, so I'm assuming you're gonna show us something. Uh-oh, we have a demo today? Oh, oh, yeah, yeah. I got it. I'm ready for it. All right, perfect.
And just for a little bit of housecleeping, housekeeping, excuse me, I will be keeping my eye on the Q&A window. So if you do have any questions, go ahead and throw them in there. I'll try to ask them, you know, and interrupt Matt as he's going along. Or, of course, if we have time towards the end, I can ask them there.
As well as if, you know, just say hello in the chat, let us know where you're from, let us know if you're, you know, if it's early enough or you're still drinking your coffee like I am. But with that, Matt, let's just get started. So Matt, I know you talk about this all the time. Maybe folks on the, in the audience or on the chat aren't as familiar with VCF, which might
be hard to believe at this point after two years of Broadcom's acquisition and talking about it all the time. Yeah. But tell me, like, what makes up or what's the fundamentals of VMware Cloud Foundation? Yeah, you know, it's in a lot of ways it's changed, in some ways it hasn't. I mean, we both have worked on this product line even in previous lives.
So, like, some of the challenges with VCF is it, it has baggage, and I don't mean that in a negative way. I mean, it has a history, right? And products change over time. So, like, let's talk about where we're at today because there are some pretty big changes in the 9.0 track of VCF.
So one, they have a concept of a fleet, right? So you've got VCF Ops and automation at that. Have you heard of VCF Ops? I hope you have because that is now where licensing is based out of. That's the number one thing that people should be thinking about when they're moving to 9 is
VCF Ops has to be in the picture somehow. But that, that's at the fleet level, right? So those are kind of like our management areas for the entire VCF, environment, right? And then we have sub-domains in these called VCF instances. This is the stuff we're more familiar with, like vSphere, vCenter, you know, NSX if you're
using, NSX today. but you have to understand that there's like a fleet and then an instance level. Now, you know Langer really well. VCF new It may be new to some people. It's definitely not new to us, right?
'Cause we're across the entire stack in this. So, like, whether it's FA management packs for VCF operations, FA content packs for automation or workflows for automation, what we'll actually see today as part of this engineering effort that we've had since 2012 with VMware is the vCenter plugin. we were one of the first ones to make a plugin, for storage.
And think about it. We only had a workable product in 2012. So since our inception, we've been working with VMware in some flavor, right? That's something to take away. most everything you're ta- we're talking about with EverPure is the FlashArray family, right?
So these are the dual controller systems, the scale-up systems that, you know, all-in VME that drives, the lowest latency and the highest amount of, of throughput and IO, for virtualization workloads, right? Which tend to need that type of performance and, and that IO profile. And then we have so many other things that fit across the, the VMware, stack that may not
necessarily be direct VMware integrations but flow into, like, a software-defined data center or an enterprise data cloud. for example, like Fusion, Pure Protect, Pure1, and our Evergreen model, which is a bit of combination in a business model. If you've ever worked with us, you know what that, that technology is.
Yeah, I like to say, you know, I didn't highlight them on the slide, and I probably should. A lot of these are what I refer to as free gift with purchase. Mm-hmm. So you talk about the f- you know, the vCenter plugin, these management packs. You know, if you're a EverPure customer, you get, you get those, right, when they come with
your FlashArray when you set it up. Pure Protect and Evergreen, you know, there's different Evergreen models. That's kind of a, a consumption-based purchase model or an SLA, but, I mean, there's different ways to acquire a FlashArray from us. And then, Pure Protect, which we need to update this slide, Matt.
It's been rebranded. But that's our disaster recovery as a service. It's, it's now called EverPure Data Protection, I think. I- Right missed the name. They just changed it. But I just wanna call that out that, yeah, some of these on here are free gift with purchase, I like to say, but as well as
there's just dif- things on here that we would get into in a different webinar that make EverPure different on how you might protect your workloads or even procure the FlashArray in your system or for your environment. Definitely. Yeah, so, you know, one of the number one things that I mentioned that there's some history with VCF, and, and Langer knows this
as well as anybody. you know, what I will What I want you to take away from this webinar as it pertains to EverPure and VCF is that as long as the protocol and the array are on the HCL for VCF 9, you can get there as a storage top. I just like, a storage type.
I just wanna stop there and say that first and foremost. How you get there may be different deploy- depending on how you deploy.Right? So that's the point of this flowchart, right? Again, just take away whatever is on the HCL and is a supported protocol will work on my FlashArray in VCF 9.
So let's talk about how that can vary based on how you're actually deploying it. Some customers are doing greenfield deployments, right? So they already Maybe VCF 9 correlated with a hardware refresh cycle anyway. They've got a brand-new rack with, brand-new gear with, you know, ESXi nodes and FlashArray and the SAN networking and the, top-of-rack switches.
Like, they have an entire instance that they're just standing up for VCF 9 only. That's like a greenfield deployment. If you are doing a greenfield deployment, the, the type of storage does matter on the substantiation of the system. And, and really what that is is just the workflow that the VCF installer allows among,
like, during the s- the bring up of the system. So if you're doing a greenfield deployment like I mentioned, and you get to the storage section, you're gonna be asked which of these storage types are you deploying today in this first workload domain, right? This first VCF instance.
And that could be VMFS over Fibre Channel, NFS v3, or of course vSAN. Those are the three storage types that are deployed. But once you get through the VCF installer and you deploy this instance, you can attach whatever, protocol you want, whatever storage type you want as long as it's on the HCL, right?
So there's a lot of confusion here of people see this greenfield deployment wizard, and they see that it's, like, Fibre Channel and NFS v3. So we do get questions like, "Is iSCSI going away?" Or, "Is NFS 4.1 going away?" No, it's not. It's, Or NVMe, "Is NVMe not being invested into anymore?" No.
It, it is absolutely still there. It's just a workflow thing. Now, if you do a converge brownfield, which there's a couple ways to do that now, you could use the converge process or a more manual process. As long as you have VCF ops in the, in the picture, then wh- whatever goes.
You know, whatever's on the HCL, is supported, and that's Fibre Channel, NFS 3 or 4, iSCSI, vSAN, or NVMe over Fabrics. so again, this product has some history, as it pertains to what was supported in the past. It's actually way more flexible than ever. In the earliest days of VCF, it was vSAN only everywhere.
Like, that's the only thing that was supported. That's not the case today, and there's, plenty of VMware blogs you can read through, too, that talk about this big change. So- Yeah, and the primary storage on the new deployment, you know, I want folks to think about it as well as it's what VMware can automate in the bring up process.
Yeah. So this is what's supported for now. Now, I'm not Don't take this to the bank, but that doesn't mean they won't support other one, other protocols in the future. Again, you said the word workflow, Matt. I just wanna hamp- hamper on that where it's like it's not cloud, cloud builder, but
whatever the deployment process is, whatever VMware can automate to stand up that- Yeah initial management domain, that's why these are the primary storage options. Yeah. So, if One other thing I would like to ask is if anybody wants to put in the chat, is anybody actually m- is planning to move to VCF 9 or has already moved to VCF 9? I'd be interested if you wanna throw that in the chat just so we kinda
know what folks are doing. but Matt, you mentioned NFS. Yeah. And you know that I've been a longtime VMware NFS guy because of some management stuff like that, and I was so happy that FlashArray launched file, you know, two years ago now.
But kinda give us a quick under-the-covers look of using NFS or what our FlashArray file services look like if I was gonna deploy NFS for VCF. Yeah. No, exactly. You know, some people are rethinking, you know, 'cause there's, there's some reasonings for if you, maybe you're a vVol customer and coming off of vVols and you like the per monitoring,
we'll actually see a demonstration of what that means for NFS, you know, statistics. You know, some people are, are reevaluating how they lay out their data stores and what data store types are, and one of them is NFS because of this exact thing. Well, if I gotta do greenfield NFS, maybe I just start moving to NFS. So there's all sorts of things, why people are looking at NFS on FlashArray, let alone one of
the great questions here in the chat already, is when will NFS work with Active Cluster? Linger, keep me, keep me honest on this. Active Cluster is announced. It's not quite there in the code, and then, we, I think we're looking at support for that, like, the third of this year.
Like, um- Or quarter, yes or like the third half. So we're, we're really close to that. That's why we wanted to talk about this, which is so exciting, is, like, this has been one of the stopgaps of, well, you can do async, but you can't do, like, active-active setup on NFS. That's a non-starter for some environments.
That is so close on the precipice right now. It's almost there, right? Yeah. So sometime this year you will see that happen. Yeah, yeah. And it is coming. stay tuned to, you know, our website and some other news that we got. Yeah.
It, it, it's coming very soon. So. Nice. Yep. So, you know, if you're looking at FlashArray File, maybe you're looking at, adopting more NFS workloads, for FlashArray specifically as it pertains to VMware, there's a, a number of objects you have to be familiar with, and if you've worked on other systems,
they're fairly similar. So you have the array servers. Array servers are ways to kind of, like, separate different file sy- like, file environments. Maybe you wanna have it for a prod environment or a, a dev environment, or maybe you've got
servers that serve, you know, data or NFS shares for, like, departments, not necessarily for ESXi hosts. The servers are a way to separate that, so maybe you can set them up to ever other LDAP authentication sources. that would be, like, one example.
So there you have the servers part of it, and that's one thing to keep in mind. Then within servers, you have file systems, and the file systems are gonna be essentially where your directories live and, where you actually map the exports to your host. And you could have multiple file systems within array servers, to serve different purposes.
And then at the end of the day, you have the policies. This is where you would have your export, policies, which dictates things like, am I authenticating via Kubernetes or is it just, you know, no root squash, those types of configurations for the export. You can apply quotas, which can be really important.
Like, you know, the default when you go make an NFS data store, I think it says 60 petabytes. It's a arbitrary large number 'cause of the way the NFS protocol works and how ESXi works with it. You can add a quota to make sure that it actually shows the storage you want to show.
So maybe you want it to be only 20 terabytes or 60, you know, 60 terabytes, whatever that number actually it is. You can do that with quotas, refresh the data store, and you can get thatAnd then, you can have audit for access logging so you can make sure that you have some sort of, like, security process to go pull logs on who's been accessing this.
And then we have auto directory, which is a really cool feature that we'll look at in this demonstration. All right. Yeah. So let's go to- Yeah, just one thing I gotta I need to show my age for the audience is when Matt first talked to me about array servers, I was like, "Oh, it's wack wack server-prod", and Matt goes, "What's a wack
wack?" And I was like, "Oh, man." Yeah. Never mind. Never mind. Yeah. as well for myself, if you guys are interested, like, how all this gets built out and formed, go out to everpuredata.com/demos, and I've actually got demos of building all this stuff out if you're not familiar with it.
But had to tell the wack wack joke, Matt, sorry. Yeah. But moving on. No, absolutely. Two, two generations here. We got a Gen X-er and a Millennial, so you know. Yeah. okay. So VAI features, if you're gonna use NFS,
you'll want to make sure this VIB is installed because you get access to VAI functionality, which takes a little bit more than on Block. You need a VIB on the host, and fortunately with VLCM it's way easier than it e- ever has been in the past. We'll show you what that looks like in vCenter. We also have a KB article of how to add this.
Either maybe you don't like to do VLCM, you like to do the manual, e-, ESXi method or PowerShell, all, both of those are gonna be supported. But that will give you full file clone, fast file clone, reserve space, extended statistics which we'll dive into today, and then on 9 it'll be unmap. So if you go to VS- VCF 9 or vSphere 9, unmap is now available, and in the 1.4 version of
the VIB, u- unmap is now supported for NFS. Yeah, and this is a, a, a key differentiator for those. I, I saw that somebody was using VMFS. I think it was John, so basically Block. This is a key difference unfortunately from the VMware side where VAI is on an E6, except
for Block is it's there- Mm-hmm for lack of a better word, right, Matt? Like, it's, it's- Yeah in the kernel or whatever. Yeah. Whereas NFS, you do have to go through the extra step of installing it or adding it to your image. So that's just one thing to, to clarify is, like, it won't be it's not baked in, right-
Yeah like Block is. You do need to do an extra step, so. Absolutely. So we're just gonna go over these next couple of slides real fast, just to give you a basic understanding of how the hardware's built to run this service, and it's just good to know before we jump into the demonstration. So first off, it's a first-class citizen.
It's not a VM on top of Purity. It's baked w- as a service within, Purity. So this was built to be a first-class citizen, for this because we are a multi-protocol array and we wanna give all services equal, functionality, and that, that's why we built it the way we built it.
It is an active standby pairing, right? And that's more of how NFS works, and that's the reason why we're looking at, like, how this is set up on FlashArray, which is slightly different than Block. All front-end interfaces on Block are active, but still the evergreen model is that one there's one primary controller running all these services so that that way if you fail
over, you don't get in a situation where you ran your controller or your dual controller system at 150% and you failed over, now you have reduced services, right? So this is, like, slightly different than Block just because it's a different protocol, but still the same evergreen model of ensuring that we are able to go remove these when you have something like a evergreen renewal where you get fresh hardware, right?
And you can do that during business hours and non-disruptively. So dedicated NICs per controller just like any other service. and so we're looking at, like, a primary controller, file services, and block services. We've got eth0 and eth1. Yeah. So- And I on this slide I should, probably
should have updated the comments. Two is minimum, right? Mm-hmm. You're not bound to just two. That's obviously the minimum for, you know, redundancy, HA. but we do support more than two interfaces if you need that type of, you know, bandwidth or whatnot.
But it's two's the minimum. I'll, I'll update the slide, Matt. That's kind of a bad note there, but- Yeah just know if you need more, you can do it. Yeah. So obviously since two's the minimum, if we have a virtual interface go down, you know, w- you know, we'll fail over to eth1, right?
So very simple. This is why we're required to we have, you know, still a virtual i- in- the interface can move between these two physical interfaces to keep services up. And then if we go and look at, you know, the, the next scenario, which is like a full f- controller failure, that of course would boot the NFS client over into the secondary
controller, which will have a, a, you know, a pause in IO, but will resume IO as soon as that takes over. so again, nothing, nothing groundbreaking here, just so you know how, you know, this works, how we fail over betren- be- between controllers, how we're a, a highly available system.
We have dual interfaces. We have two controllers, and obviously we have RAID HA, which is what's going across all these drives. So if we have drive failure, we can also recover from that. So- Perfect pretty basic stuff.
Yeah, everything I would expect to see, right? Yeah, exactly. But just to show everybody it is there. This is how it works, so. Yeah. question- So Oh, sorry, Matt. Yeah, a question in chat before we even, get to demo time.
some of John asked an opinion on, more storage vendors or vendors moving to NFS instead of Fibre Channel. Is that what you're seeing as well, and suggested probably look into ge- getting off Fibre Channel and move over. I would say the, the conversation isn't so much Fibre Channel to NFS as it is Fibre Channel to Ethernet protocols.
you know, NFS is still, you know, It's a protocol that's been around a long time, so with that comes, both good and bad. Like, it's got some, you know, hi- some legacy components to it, but we i- iterates with 4.1. Matter of fact, one thing we didn't discuss is, like, with NFS 4.1 and the VAI primitives, like, the Block and file world are getting more overlapped.
If you look at the Venn diagram where things were misses for NFS and misses for Block, they basically, overlap pretty highly.Um, but at the end of the day, I think where the most of the market is talking about is just Ethernet-ba-based technologies instead of fiber channel because pretty much all the hyperscalers have no fiber channel in their environments.
It's all going Ethernet. And I was just in a data center recently and seeing these 800 gig NVIDIA Mellanox switches, which far outpace what s- fiber channel's doing. So it definitely seems people are going to Ethernet, so NFS just falls into that bucket. Any, any thoughts, Lingar? I know you see this, too.
Yeah, I, John, I hate to tell you this. Is, it's depends, is what Matt just basically said for two minutes. to Matt's point though, Ethernet for speeds is cheaper. You know, fiber channel is fraction of the speed that you get with Ethernet. But remember, it's latency-specific for your workloads.
So while if you have highly transactional workloads or maybe you're doing stock trading or something where you just need that latency of fiber channel, fiber channel's going to be the right fit. If, if you don't, maybe Ethernet's the right fit because it's p- potential cost savings or whatever. But you need to look at your applications and kind of figure out what those latency
requirements 'Cause that's really the key difference between FC and Ethernet. Ethernet can be super fast, but can also be a little slower as far as latency is concerned. Like, let's not confuse the two. So, um- Yeah that's always been my experience for p- for a long time, is like really let the workload dedicate what you're trying to do.
and then maybe budget comes, you know, maybe option number three or four down the priority stack, so. Not to mention that this- Hope that answers for you, John. If not, throw something in the chat and we can kinda keep going back and forth there. And, and it's a hotly debated topic, too. Right.
I'm sure everyone's got an opinion on it- Yeah but that's kind of our take. That, that's my way of putting it back on the person, is like- Yeah you tell me what you think is the right one. I don't wanna have- Yeah a religious battle with you right now, so. Yeah. We could ta- we could talk about that for another 45 minutes- Yeah there's no doubt.
But, um- All right. So Matt, I think you've got a demo for us, right? You did this l- you're gonna do this live? Yep. We're gonna do it live. All right. I like it. Oh, boy.
Let me stop my share, and then we'll turn it over to you. Yeah. So give us one second, everybody. All right. Let me know when I am live. And you should see a very familiar interface. I see it, correct, yes.
Cool. Everybody else does? Throw a Y in the chat if yes or an N for no. Yeah, definitely hit us up in the chat. I'll start talking if, uh- Yep you can pause me. But, so, you know, vCenter, everyone's familiar with this.
The first thing I'm gonna note is this is where you'll find the VAAI VIB, that you would go download. Again, if you're gonna do NFS, this is highly recommended for all the capabilities that we mentioned. Full file clone, extended statistics, all, all of those things that come with the VAAI VIB, you can download it here.
Notice the version support. So the unmapped stuff is in nine because it's a nine feature, right? So, keep that in mind. How do you upload it? the easiest way today is really just doing it via VLCM.
And so this is what's replacing VUM. So if I go over to my cluster level and I go to updates and then image, and I go show these additional components here, you'll see that we have some Cisco UCS add-ons. And when this f- fully loads, you'll see that we have the 1.2 version. We're, we're a rev behind in, in this cluster, but, this is where you would upload it.
So you can upload your, the Download the VIB, upload it, and then you can go through and remediate the cluster to go ahead and get it updated on each house. That's the easiest, GUI way to do it. There's stuff in the, the same KB around doing it with PowerShell or ESXCLI. Whatever fits your fancy, whatever you're most comfortable with.
This is probably just the more modern way, to do it in vSphere. So let's go ahead and set up an NFS mount. The best way to do that is the plugin, is the vSphere plugin, and I'm gonna show you why. So the plugin's already installed here. If I just go right-click on a workload cluster and I c- create datastore,
I'm gonna choose NFS. I've got all these options. I'm gonna choose NFS for this deployment. I've got the workload cluster object that that's where I wanna connect it to 'cause I want it to be available across all my hosts in that cluster for HA purposes and moving, VMs around.
I'm gonna pop this guy on, one of our arrays called SN1X9DR2F0533, and then I give it a name. We're just gonna say expert-led demo. I'm gonna leave the default of one petabyte. You could go through and change that size or even add a policy within, your, your quota policy within Purity to make sure that that shows
lower, but we'll keep that. this is a, a, a fantast- outside the ease of workflow that I'm gonna hit, throughout this, this is another big reason to use the plugin for NFS deployments. here we've got the inConnect settings. So this is another reason why NFS is getting closer and overlapping with the performance of
Block, is because of things like session trunking. So if you ever go deep into NFS, you know, there's basically, these connections that are set up but only go over one flow, which kinda hurts the throughput of NFS in some ways. So usually you're trying to fix that at a higher networking level with LACP and other, other ways. But now with session trunking and, and
inConnect, which is this setting specifically, I can actually have multiple TCP sections in that one connection to generate more throughput when my VM workloads are running on this data store. The recommendation middle of the road is four. That's what I'm gonna do here. but you could go up to eight.
Just know that each connection takes a little bit more CPU power from the host, so there's a balance here of, you know, s- CPU load on the host versus throughput considerations for your, for your disk usage on your disks. this is a, this is the VIP that we're talking to on the, on the array server. So I've only got one on here.
Makes it pretty easy to choose. In a production environment, you would typically have a VM nic, a VMK that would be associated with a VM nic-Right? And you wanna bind it to that just like you would do any other storage traffic. This is the thing that gets most, the most often is you still want a storage area network
that can be still embedded on the top, same top of rack switches, but should be on the same layer to domain. and you should be binding it so you can have dedicated performance. Yeah, logical separation's still, still a good idea. Still a thing. Does- Yeah, still a thing.
Yeah, that doesn't go away. Yeah. Exactly. n- and network design matters. the pod we're not gonna put it in a pod, but definitely if you're looking at active cluster, you would be putting this into a pod, would be one of the things that you're doing. And then, this is the export policy.
Again, another benefit. I'll go show you what this looks like in the FlashArray, but we're gonna do unrestricted. And then all of this is the information that we have, and I just click Finish, and it creates this datastore. Now, I put this on FO533, and I do have that guy here.
Let's go pop into him. Which is the FlashArray, by the way, since I don't know what that means. yes. Correct. The, the, RX90, FlashArray. Your acronyms mean nothing to me. I know. We start talking our own shop, and it's like
it's assumed. Yeah, no, this is an X90R2 running 693 is what this is. And, you can see the FlashArray here. That process did finish. You can see the Create Datastore. So if I go down here and look for, our expert-led demo d- right here, we've got it.
It's created, and we can see, it's, it's ready to take our virtual machines. so if I go back, to the FlashArray that it's actually deployed on, which was this FO533 which is an X90R2. If I go over to my file system, so I can just search that name, and you can see basically all the things that we'd have to do piecemeal inside of Purity, it the
plugin did in one workflow. So if you did this yourself, you would need to make sure that you created a directory export, you assigned those, you had the policies created. So one of them here would be the, I believe, it's the ex- expert-led demo export policy is created.
You can see it's for that. If, It also automatically creates an auto directory policy, which you can see here. And again, the, the auto directory policy makes a managed directory for each virtual m- machine. Now, if you're in the low thousands of VMs, there is a high ceiling here, so this may not
be the best, feature for you. But if you're not in that area, I would highly recommend using auto directory because this is the last thing I'll leave you with 'cause I know we're getting close to time. When you do NFS with a V- VAAI VIB and you have auto directory enabled, which is default with the vSphere plugin, you get extended statistics on a per VM level.
Now, I don't have a whole lot running, here to necessarily show off, but I can show that chart right here, which is you can see, this, this, VM here is on a NFS datastore. It has, auto directory on that, NFS datastore. And if I go to the Performance tab, I get a per VM number instead of the datastore level.
VMFS is a clustered file system, so there's h- no way to really peek into each of those and get performance metrics on individual storage performance for that. But when we make managed directories, we can actually see the statistics on each individual directory for that virtual machine. So I can go say, "I'm having a problem with this, you know, application." I go look at it,
and I can see specifically what's that VM experiencing from a latency and I/O bandwidth perspective. So hopefully this gives you- I can't Sorry, Matt, just real quick. Yeah, yeah. I can't understate the value of this as, again, as a guy that started his VMware career on NFS- Yeah a long time ago.
The value in this is super great because, again, for that meantime to innocence or that troubleshooting, you know, even if you see it on VMFS, right? Like, that's what VVols tried to fix is like, "Oh, I've got all these VMs. Who's the noisy neighbor?" Where now you can actually go in and drill down on the VM, and I just cannot highlight how cool this is from an NFS datastore perspective in the, in the
VMware environment. So thanks for touching on it. It was, it was a lot to consume, but the, the value can't be understated for that, Matt. Thanks. Yeah. For sure. But that's, that's all I had, you know, for this session. I know we're getting close to time.
you know, big takeaways, you know, again, VCF 9 on, on FlashArray. We're multi-protocol. You get to choose your protocol as long as it's on the HCL. That's one of the big takeaways there. If you use NFS or are looking at using NFS on FlashArray, it's a great offering.
It a- has VAAI functionality. You want that VIB on there to get that functionality. And specifically for those that maybe are really sad, like I am, that VVols is going away, if you want that per VM statistics that VVols did give you, NFS is a great option here. Like, and something strongly worth considering.
Yeah. Matt, we did get a question that I was g- holding off for you. Um- Mm-hmm Levi, hopefully I'm saying that right, asked about how many Pure customers I'm gonna read it verbatim. How many Pure customers who are still using VMware today do you see who are upgrading to VCF or vSphere 9 but are not converting their
VMware installs to the VCF, VVF kinda nomenclature? Basically, I think what they're asking is not using, like, SDDC Manager and ARIA Operations. Yeah, that's a great question. so I will say that this, this iteration moving to 9 is I think we all knew was gonna be slower than the rest because of these big changes.
and, so I'm seeing probably, o- one of three areas. I'm seeing the greenfield, the converge method, and then your method you're referring to, which is I still deploy ESXi, vCenter, and then get it into, the state it needs to.I do see that happening. So it's not, like, technically VCF, but it has to have ops.
You have to have ops to validate the keys. And so, like, at that point, is it, is it VCF or VVF, or is it not? VCF ops is there no matter what, otherwise you're only valid for 60 days and your stuff's gonna go disconnected. VCF ops is the only way to valid- validate the keys.
So in that sense, I would kind of maybe reframe the question of more how many people are doing it the manual way. I do see that in the field. I would say it's on, the lower third, but, you know, we still have a ton of 8 customers going to 9.
That could change. That could be the more common approach. a lot of people are starting, like I think I saw in the chat, where they're getting ready on Greenfield first, just to get it filled out in a dev environment or whatever the case may be, but of course you gotta make sure you got the hardware to support that. Yeah, because, A, correct me if I'm wrong, is it October of 2027?
You got it, man. Yeah. We're, we're, you know, we- It's one of those things where we've got time, but it's gonna come up faster than we think, so. Right. so we're still in that kinda bell curve of, like, we haven't even hit the bell curve of the biggest chunk of customers moving to 9 yet.
Right. And Levi, thanks. Looks like we answered your, your question right, so thanks for putting that in the chat. Yeah, great question. we'll wait for another minute if any other questions come in, but I do wanna mention we have our Pure//Accelerate conference coming up in, jeez, like 20 days from now. Yeah.
That's coming up fast, Matt. Get your presentation done, Matt. I know. this is our EverPure Conference. This is a free conference. There is no cost for admission to join.
You can come see Matt, give him a hard time, meet PMs, learn about our latest products and services and whatnot. obviously it does cost you a airfare and a hotel, but this is a two-day event that we put on, bring out, you know, roll out new features and functionality and talk about all that. So if you're interested, scan the QR code to register for that.
And of course, if you're not, go ahead and sign up for our EverPure Customer Digital Community or our communities page. This has been relaunched within the last year. New interface, better, you know, features and functionality. And this is a g- spot where you can, you know, ask questions, but what I say that is don't
ask, like, tier one or PRI One support hard down questions. But if you want some fe- questions, features, functionality, post the d- questions in there. Myself or Matt or one of our other folks might answer. But again, you know, if you're having a, an outage problem, please do not post it here. Call your support people as normal.
I didn't see any other questions coming in. I don't think so, yeah. So I think with that, Matt and everybody else on the line, thank you for joining us. I hope you found this a helpful session. We try to do it in 30 minutes or less usually, but Matt might get a
little verbose as you know. Yeah. but- That's pretty typical. Right. Thank you for joining, and I hope you have a r- fantastic day and enjoy the rest of your week. Until next time. Thanks, Matt. Have a good one.
Matt Webb
Senior Field Solutions Architect, Everpure
Jason Langer
Senior Technical Evangelist, Everpure
Join us for this Expert-led Demos webinar and learn how Everpure FlashArray™ simplifies and accelerates VMware Cloud Foundation (VCF) deployments by serving as a high-performance NFS-backed datastore.
Whether you’re planning a new VCF deployment or want to optimize an existing one, this session will give you practical guidance you can apply immediately to simplify deployment and lifecycle management.
This integration can:
- Streamline Day 0/Day 1 setup
- Decouple compute and storage in VCF for better scalability, right-sizing, and improved cost efficiency
- Deliver predictable performance and data services for your cloud foundation
- Enhance protection and resiliency for VCF workloads with built-in FlashArray data services
We Also Recommend...
Personalize for Me
My Updates
No updates at this time.