00:00
Welcome to accelerate. Hopefully you're in the right uh right session and end immutability with VM and pure storage. If you're not, please don't get up and walk out. No, I'm just kidding. Uh My name is Chris Brigg. I'm a principal technologist for pure storage. Uh Actually just recently became a principal technologist, been doing modern data protection
00:21
for the last two years uh at pure storage with our, our, our fine friends over at uh with me today is Mark Pullen. Introduce yourself. Yeah, thanks Chris. Uh My name is Mar Poland. I'm a solutions architect with VM and I am the technical lead with the Beams Pure Storage Alliance.
00:42
And last but not least our, our guest of honor is Chris Weis today with RC Willy. Um Chris, you like to say something, Chris uh Chris Weiss with RC Will. We're a home furnishings company and we are in about four different states. But uh I am a senior system admin on the infrastructure team. So I do all things beam and compute and storage and that's what I'll be talking about today.
01:08
Thank you. So excellent. Let's get going here. So just a quick agenda. You know, we've got Chris with us, we know that um you know, hearing from us is one thing, but hearing from a customer that actually does, it has found a way to really utilize our technology together jointly pure and is,
01:26
is really what we want to hear. So we're gonna let uh you know, Chris share his story to start out, we might interject a little bit, ask a couple questions and then we're gonna talk about what's new with VM 12 and, and us API version two that beam 12 came out earlier this year. February, I believe. And uh we've done a lot of joint engineering
01:44
together, so we'll, we'll touch on that and then lastly kind of wrapping up the whole, the whole story is the end to end i immutability that we do together, uh that we'll talk about, which is obviously the session title. So, uh with that, I'll, I'll give it the stage over to uh Chris and uh show us all the awesome stuff he's, he's doing.
02:03
So how many here have a backup solution currently in place by raise of hands? OK. How many here know that your backup solution is a restore solution, meaning you can test it. It's verifiable and that your backups that you did last night can be recovered wherever you need them to do. Show of hands. OK?
02:27
That's very impressive. And that's what we're going to talk about today is V is a great resource. And we've been using it for a number of years. I started with beam back on version 6.5. And so we've seen it grow over the years and it's got a lot of technology. However, what we ran into as RC will, we realized I've got this great backup solution
02:50
because I had this data D appliance where I could store tons of data over and over and over. We had 120 180 days of backups. And then somebody said, well, can we recover those? And I said, well, let's find out. And if you've ever been in this situation, what's your biggest challenge?
03:12
What's that? Well, space time. Yeah, because you have to rehydrate all that data if you want to recover all that data. And for us, that was between 24 and 36 hours. So we would restore all that data, then we would run these powershell scripts that would take it and it would move it into our V center. We'd register all the VMV
03:38
mx's and then we could turn them on. But then unless I wanted to go through every single little machine, I couldn't verify that they were good. I'd say, well, they came on, we're good. And so that was our challenge. And so when we realized that we had a restore
03:55
problem, not a backup problem, and we needed a restore solution. We started looking. So we, we started talking to our current vendor who was supplying the data uh appliance. And they told us, oh, well, we can do this. We realize that the, the model you're on doesn't do exactly what you want,
04:15
blah, blah, blah, blah, blah. Another thing for us was immutability. We needed to know that our backups were going to be safe from ransomware from attacks by bad actors. So pure came to us as one of the options and we started listening to them. Now, one of the things that we had with our current is we needed the ability to at least
04:39
get 30 days back up with their. And then we needed to know that we could restore quickly and we wanted to, if possible utilize sure backup, how many here have ever used? Sure backups or virtual labs, which is sure backup. It's a great technology. If you're not sure what it does,
05:01
it can take your existing backups, it turns them on, spins them up right off of storage right off the backup and then you can run all sorts of tests against those. You can even leave them on and run them in an environment where you can do testing. So she backups and virtual labs is a great technology.
05:20
The problem is when I tried to use it on our existing environment. After 30 minutes, I still didn't have a single VM that was turned up, it just didn't work. So working with our, our, our, our peer engineers, they ran some numbers and they said, well, these are the numbers that we're getting.
05:40
We said, OK, well, we're gonna give you a try. So we took it on a 30 day trial and we brought it in. And as you can see, this is our environment. I had uh we have our Dell that the Power max is our primary storage. We're not even using this for the flash array C for our primary storage at this time. And so then we have our VM back up server, I back up the data and I put it on
06:07
to the flash array C as the target. Then we started running some tests and said, wow, I'm able to spin up a machine on an instant on in 2.5 minutes. What I couldn't do in 30 of our existing appliances, we were able to do off of two or 2.5 minutes to three minutes of machines.
06:34
That was impressive. And so then we thought, OK, we got to do replication because our current appliance does replication. So we set up the replication and we started moving it to our dr site. And I said, I wonder, I'd really like to know if my backup was not only good,
06:52
but my replication is good. And so what we did is we developed and you can do this if you have two flash A A CS doing the replications. I have a backup manager server sitting up in our DR site and it just goes out and it looks at that repository and it says, huh, what data do I have out there?
07:12
And it catalogs it, it's not even aware of our environment down in production catalogs all that information in and then I ingest those into these sure backup jobs that I've created. And in our dr side, I don't know how your DR sites are. We're more active, passive where it's just sitting up there. So I have all this data or all this compute,
07:37
I should say sitting up there doing nothing waiting, put me in, coach, put me in, you know, so I thought, OK, well, how can I make it feel relevant in the little it world that we live? So I utilize that every morning, it picks up my backups. I run it through, it runs through a sure backup on every single VM that we have and then I get
07:59
a report. So I, I check about between five and 600 V MS a night were between the backup, the replication, the rec cataloging and the restore and all this was made possible because of the marriage between VM which had the technology and pure, which gave us the infrastructure and so doing that. Now, I have a report every day that tells me which machines are verifiably backed
08:29
up and recoverable because it brings it up. It does a heartbeat, it does a ping test. I can give it other tests. If I want SQL tests, I can run scripts against them and bring them in. Now, where does this also take us? Well, we now have a recover solution. But the other day I was tasked with,
08:48
ok, I need you to write a powershell script, but we need to take all this data from hr and create user accounts for all these users that don't have one in active directory. Well, I don't want to just run that against my production D CS, right? That's kind of dangerous. I said here's what I'll do.
09:09
I'll do a backup of our DC. I backed it up, took a couple of minutes to back it up. I created a virtual lab. I brought it up in my environment. This direct, this domain controller is only a few minutes old from its original state. And now I can run anything against that in a virtual lab and the marriage between the two
09:32
technologies makes that possible. And I'm actually using my domain controllers to be able to do that. And so for us, this is where we were able to see speed, unlike anything we expected it exceeded our ability or not our ability, our expectations. I'm sorry. And security, we've turned on safe mode and with safe mode,
09:55
if you're not familiar with it, it makes it so that if a bad actor were to get on and delete a snapshot that we've created, they can't annihilate it. All I have to do is say, bring it back and we tried this and we tested it and now we've got these things sitting out there that we can't get rid of without calling support. So, uh as far as simplicity goes, it was actually pretty simple
10:22
for us to figure out with a few helps from our pure field agents. They were able to say, well, this is how this works. And then one of them was very excited, one of our field guys. And he was like, hey, um did you know that you can do envy me over fiber? Has anybody ever done that? OK.
10:42
Well, we have that with our peer and we're like, well, let's turn it on. We don't need it, but let's turn it on. And so we did and I don't know if you know what NVME over fiber gives you. But in a standard storage array network, your scuzzy cues are 32 deep. And then, so that's your max Q depth that you get.
11:03
Well, NVME supports 64,000 cues and each queue is 64,000 deep. So it's exponentially just huge. And so, and you would think you're now going to get a bazillion times more in data, but you're not, you're only gonna get, we got about a 30% increase. So, but it was there, we turned it on.
11:25
And so that speaks to how pure has divined their interface that allows us to be able to do that kind of stuff. Scalability. We haven't had to scale out. One of the things that we've seen is our, we set our, our backups at 30 days and said, let's see if we can get 30 days of backups when we're at 25 days. We were still just utilizing a simple portion
11:49
because of the, that, that pure does. And we said, ok, let's move it to 60 days, let's move it to 90 days. And at 90 days right now, I'm getting 90 days of backups. Um We're at 37 to 1 on our d which blows the socks off of what we moved from. So one of our biggest,
12:13
yeah, one of our biggest fears was just not realized. And so we use it also for some vmfs volumes that we pointed to for some, some of our applications we're utilizing just this little Pure box in ways that we never thought and we never could with anything else that we had had, we had a huge array of J bods, just a bunch of discs and fast array controllers and it
12:42
couldn't do anything like this. And so like I said, it has exceeded our expectations and as far as the sustainability, I've been through a number of forklifts of compute and storage and a number of other things when your support finally runs out to where it's more expensive to, to, you know, keep it than it is to get rid of it.
13:07
And the nice thing we're on the evergreen. And so we look forward to the day where we just upgrade, they just upgrade us. We don't have to do this and we know what our spend is going to be going forward. And that's really nice. Any questions for me before I hand it back to these guys? That's, that's my story.
13:29
The bigger your back four days before they did do, we're backing up. I want to say probably about 50 to 60 terabytes a night somewhere in that range. And the one nice thing about this is this the pure storage and V is also backing up our physical servers running agents.
14:00
So all our red hat stuff, all of that is getting backed up on there and we're still getting the d that we're getting. It's not just like we have hundreds of the same kind of machine. Any other questions before I hand it back to them? Are you a it's all automated through powershell? You guys handle the rep, you guys handle the replication and then I use Powershell with VM.
14:30
Yes. Very good. No, and we use that. We do. Yeah. Barks. Loves it. Yeah. No. Anything else. This is something that you should at least look into if you've got pure and you're using VM,
14:54
start to use its capabilities because you can open up the world, you can check your backups and I believe Zain was putting together the scripts that I wrote in kind of a generalized format and he was going to make those available. So I have a couple of questions for you, Chris. Oh, there we go.
15:29
Currently, my backup environment is on a work group. It's not part of any domain. So and I'm hoping to move that to a completely isolated V land. Um We also use tape so we do our air gap solution to tape. And yeah, we have four or five different parachutes depending on the tier.
15:51
We're backing up. We use V MC BP. I don't know if you're familiar with continuous data protection. We use that and we actually go from our power max on prem to our sea. I'm not using it to go for AD R solution. We're using it so that if we have a problem
16:11
with our tier ones in production, I recover off the power or the pure and it's fast enough that it it does, it just fine handles it just fine. You guys had questions. Yeah, I won. So it was a perfect segue that question. Um You know, one thing, Mark and I even discuss this, you know, looking at your environment,
16:33
we generally don't architect two separate beam environments, right? We'll say put proxies here and there, but the way you've done this right, you have a completely separate environment, different backup server, you've almost created administrative zones, right? And I don't know if you actually have different log ins and securities.
16:52
But you could, it's kind of where my mind is going, right? You could have different admins, different logins, different credentials and different environments. And because you're using flash array replication and V has that portable data file format, you can pull it into a new backup server, right? And that's what we're doing.
17:09
Yeah, we do that and we actually have another V center. It's separate. So our DR environment is ad R environment, right? We can just bring it up and, and we do that. I can, I can spin up all these machines very quickly. Yeah, we think it's fantastic.
17:26
Anything else? Not all right to you guys. Thank you, Chris. That uh you're gonna be a tough act to follow you. It has to go. All right. Thank you. I'm not gonna step on that. Uh So I again,
17:50
tough act to follow. That was awesome. Uh So what I'd like to talk about if you noticed in Chris's uh architecture there, uh He, there's, there's an opportunity there for pure to, to get the primary storage, obviously, when it's the right cycle to do that and when they do that, they can take advantage of the version two capabilities of the US API
18:19
universal storage API capability before I get into those, let me just touch on some of the other things that we've done in version 12. One of the headline features for version 12 was promoting object storage up to what we call the performance tier. So direct to object, we've supported object storage for many years with Zoom backup and replication, but only in an off load
18:47
capacity. So you had to write to something first and then you could offload the objects storage. Now, you've got the ability to leverage flash blade s potentially even flash blade, E as that primary backup platform on prem and still do the capacity tier off to another object storage platform if you so choose, could also go to other platforms,
19:14
right? No, no lock in there. What was critical with supporting object storage is the ability to have immutability for object storage and pure, like 30 days before we released version 12 released their support for S3 object lock. So they've got that immutability in the time
19:44
that's happened since the release of that release of uh flash blade purity 41. All right, I got it right. Uh And backup and replication version 12, they've gone through the VM ready uh Vaid qualification tests uh for object storage for object storage with immutability. So, uh we've,
20:10
you know, actually I give them credit, they put it through the paces. Uh Our team validated the work that they did. So, uh you know, it has uh has that validation. What to me is a critical component when you're talking about object storage on prem is the ability to protect against somebody getting the admin
20:35
credentials of rogue admin effectively somebody being able to get the administrator access to the system itself because if they could do that, then, you know, they could go in and potentially undo everything that was there. So safe mode retention lock was another critical capability that pure provided for the flash blade. And what that does is it locks down
21:07
the uh admin credentials. So that e even, or even if you compromise the admin credentials, they can't go in and destroy the buckets without uh two named individuals getting on the line with support. So uh you called that? What zones of something? Zones? Yeah. And I, I think of it as resiliency zones,
21:37
right? It's uh in my mind, I, I pictured the, you know, old medieval city with the layers of wall, right? Protecting that inner keep, right? Doesn't mean that at the end of the day that, that the bad guys can't get to it, but it makes it difficult every wall along the way. And that's that layered security model is critical these days.
22:01
So uh anything you wanted to add on that? Yeah, I would just, you know, say uh I've been getting a lot of questions of, you know, that this your solution together already worked well, right? And you're using flash blade and NFS, why, why object, right? And, and a couple of reasons because it's
22:18
becoming ubiquitous, um It actually simplifies it. Now we get to utilize uh S3 object lock like Mark said, which means instant immutability. As soon as the objects are written, they're immutable, right? We're not waiting to take a snapshot and protect those. It's instant we no longer have to,
22:33
to scale out connections for NFS, right? The way beam handles S3 and flash blade, it scales it out for you. So it's a single S3 end point. Now you point to one single uh uh IP address in beam check the immutability check box and, and you're golden, right? And it scales out and performs beautifully. And then we add on retention lock like Mark said, because we're not because now we're using
22:59
a standard in object lock rather than our own safe mode. We still wanted to lock down uh admin the things within the admin console, right? So that's exactly like you said, it enables us to if you, if you lose admin credentials or someone takes those or uh you get better one day and wanna go delete buckets, you cannot, right.
23:19
That's exactly what it's doing. So that's, that's why it's still called safe mode retention lock. We're actually using object lock for the immutability. The retention lock stops any admin from changing those settings or deleting the bucket itself. Yeah. And if you are considering uh object storage
23:36
for your on prem backups, uh that's a critical component. So just, you know, I'm not telling you what to do. I can if you'd like. Uh but you know, please make sure that, that is at least on your checklist of things that you're putting the vendors through the paces on. All right.
23:56
So the next one, right, a big capability for VM and pure, in fact, pure was our development partner with the universal storage API version too. What we've done if you look at the capability that was provided in the version one, it was all around beam orchestrating capabilities that pure could do on a single flash array. So we could do back up from storage snapshot,
24:25
we could orchestrate snapshots in what we call a snapshot only job keep retention on those snapshots. You know, I I always use the example of, you know, uh 24 hourly snapshots uh having your backups for 30 days, right? It just it augments that with version two of the universal storage API which came out with version 12 of VM backup and replication.
24:55
VM can now orchestrate snapshots off of that primary array, leveraging asynchronous snapshot replication to another flash array. And you don't have to have the same retention. It's asymmetric retention. So in my example, 24 hourly snapshots on the primary, then I could keep say three days worth of hourly snapshots on the remote side.
25:22
So uh and you get all the same recovery features and beam has spent a lot of time built a great, I believe a great reputation. Uh Chris validated that for us uh around recovery capabilities. And you know, you get those not only on the primary side but on the remote side as well. So the first capability, asynchronous snapshot replication, the second capability synchronous
25:52
snapshot replication. And if you're familiar with active cluster replication, it effectively stretches the cluster. So when I create a snapshot, now, vem is not actually doing that work pure is we're just asking for the work to be done orchestrating when that snapshot is created on the primary side, it is created at the same time on the remote side.
26:17
And then the the use case that you could leverage there is you could back up from that remote snapshot and have your first backup be off site, which is appealing to many customers. The third capability is to be able to offload the snapshot in its native snapshot format to another storage platform.
26:45
Uh In this case, we're showcasing flash blade object storage. Now there is a little bit of work to do to do. If you haven't done this before, there's an offload application, you need to set up on, on on the pure side, the flash a side. But once that done is done, then you can offload your snapshots. And if you're in compliance centric
27:11
industries like financial like health care and you're, you're using snapshots as part of your retention scheme, it is very likely that your compliance officer, your external regulators are going to say if you're using snapshots, then you have to keep those snapshots around for an extended period of time, 30 days, 90 days could be even longer than that.
27:40
And the Flash Blade e with that depth of storage, uh, and really compelling price point seems to be an ideal platform for that type of capability. What for you was, uh, was important about what we did with us API V two. I mean, all of that's great. You hit on it really well.
28:07
Um, I mean, lots of capabilities. I'm gonna go back to something Chris said though, right? It's not about backup. It's really about restores the these days, right? We're not here, you know, to try to make backup cooler sexier than it's been. Everybody knows knows that why we all became backup admins.
28:25
Um But restore. So all this stuff that Mark's talking about whether it was version one and doing it in a local array or now version two, doing it at local having replication. There is a feature within um in it snapshot explorers where you can restore from snapshots, right? So IV can actually crack open a flash flash
28:50
array snapshot like an egg and pull out granular granular items out of it, right? So whether I have it local or I've done asynchronous or synchronous replication and knows about that uh automatically, right? I can now restore from a secondary array at another site from a snapshot, right? I don't have to bring back the entire snapshot,
29:11
the whole volume, which would be my first step if you had a true outage or a ransomware attack and you had to bring everything back, just spin up snapshots. Right. It's instant and, and you're run off and running, you want to bring something back. Granular beam can do that from our snapshots. A single VM or a single file. You don't have to bring that whole volume back.
29:29
It, it means, I mean, for me, as far as, you know, being a data protection administrator meant everything to be able to pull that back that easily from snapshots. And that's whether VM orchestrates it or pure does it on its own, right? So if you're a pure customer today, even taking snapshots, you go install V and say here's my array, it'll recognize those snapshots and you
29:50
can start granularly restoring from those, right? So that the restore capabilities are amazing. Now, I want to make one point clear when we're talking offload, when we offload to another protocol like S3 or NFS, that's a true offload of a snapshot, right? You need to retrieve that back to the array it came from,
30:08
to, to do all this magic that we're talking about with it, right? I I think of that as an on load. Yeah. Uh Yeah. So exactly. So it just gives us there for long term. Uh Exactly. And then all those great things that uh you know, last point real quick that that Chris was
30:25
talking about, you know, he does data labs and share back up off his backups. You can do that off snapshots as well. What about the remote ones? Can you do that? Absolutely. Right. Yeah. Good question. Tricky. Um, you know, brings those capabilities right, to be able to, um, to read those and, and do those, uh,
30:43
whether it's full restores or in, in a data lab. So, what if we have, you know, use our imagination? This is not a sales pitch, Chris, this is just me using my imagination. What if Chris had a flash array as his primary storage? I thought you'd like that. Uh Then, and you could still do what you're
31:14
doing. But II I always my, everything, every architecture I start with is, you know, keep it simple, right? The kiss principal. So if you had a flash array, let's say an X at the primary site, you've got AC at the primary site is back up and you've got your remote C I can
31:36
envision where you could have VM doing backup from storage, snapshot off the flash array, X orchestrating that snapshot and, and augmenting that back up uh with storage snapshots again, I'll use my 24 hourlies, let's say uh at the primary site have three days worth of hourlies at the remote site, plus my backup.
32:06
I could even do the back up at the remote site. Uh And what am I missing there? No, I, I think those are all great points, right? I, I was gonna say whether you, you know in that scenario, whether we're using the integration or not, right?
32:23
When we're talking about having a flash rate in production and at the DR site, right? Especially with what Chris is doing. He, he's sending back ups over there. He's doing all this great share, backup data lab stuff. You could actually be sending snapshots over there as well,
32:36
right? Production snapshots. And we do have a number of customers doing this, right? And then I think of that flash ac as a true data protection platform. Now, I've got backups going to it, but I have my production snapshots going there as well. So if I needed, if I want to pull back a snapshot, I could,
32:52
but if I lost HQ or data center, one, that flash array C might not be quite as performing as my flash array X in production, but I've lost all that. I could spin up those snapshots on that C and be running my data stores natively, right? It's the same purity os we're not, we're not building a purpose built backup appliance, right?
33:11
It's we we build flash arrays to support different workloads and different needs, right? So you can get uh we do have a number of customers doing that replicating from A, from an X to AC actually. Um and being able to do backups over there or being unable to spin up there there, but you're gonna charge Chris a lot more for that capability.
33:30
Right. Oh, wait, no, it's all inclusive with the licensing. Uh also is all inclusive with the licensing. So if you license uh at the universal license level, uh you license at the workload and you get the full functionality of VM, including this advanced uh storage snapshot orchestration, primary site to remote site, offload to NFS or object
33:59
storage. Right? It's all inclusive. So it's the, the, the costs are very predictable uh initially and then over time, uh the the total cost of ownership is very predictable as well. Good point. All right.
34:16
So we've got one more thing to talk about and that is immutability. We've talked about it, I think already that you've got and does it make sense to? Well, let's touch on this, right. So, uh one thing to mention, right, flash array and flash blade with NFS have safe mode protected snapshots, this is critical, right?
34:45
They can't be changed, they can't be eradicated, uh which, you know, critical capability. So you can trust that it is the way that it was when you wrote it. Uh When does that orchestration, we can orchestrate safe mode protected snapshot. So that's critical. Uh I think at this point, maybe go to the next slide and then pull the whole thing together,
35:13
right? And you know, I I'll give you my take and then we'll get uh Chris. But when I it's, it's funny when I look at this graphic, I think this is pretty simple stuff, right? You got uh you got your pure storage for production, you got in the middle, you've got, you know,
35:30
some pure storage. But when you start to kind of peel that onion right over here, you've got safe mode snapshots and VM can orchestrate retention of those and not only just safe mode snapshots, but what we can do is add value to those. What we can do is we can que the workloads that are in those snapshots to improve the recover
35:59
of them. So now you've got uh application consistent snapshots that you're keeping retention on, on this primary site, you can orchestrate those over to the remote site. Keep different retention on them. You can keep your backups uh on the uh actually, that's not the remote site.
36:24
I I misspoke that's just the backup storage. Uh You can keep different retention scheme on your snapshots. You can keep your backups there for as long as you want as we saw with what Chris is doing, you're not limited to a single site. You can keep going to a remote site to our site. Keep that safe mode of mutability. VM can add our hardened repository of
36:50
mutability on top of the flash array. And then with the flash blade s you've got, you got object lock, you got direct object from us and then you've got your retention lock. What I miss? Yeah. No, no, that's great. I, I was gonna wait till this slide to tell Chris what he could do with the flash ray Xs in
37:12
production. Um No, you, you know, you bring up a great point. So I talked about on the last slide how you know, I, I really like to restore Restoril, right? Being able to restore, recover. Um And you say, well, that's cool Chris that so can see your snapshots and can restore granu.
37:31
But I need application consistent backups. I need application consistency like Mark just said beam can do that to snapshots, right? Pretty amazing. So now, not only do I have a snapshot, I've got the information I need to have that application consistent and I can restore those types of things really cool. Um You know, the other thing I would point out here, uh I mean,
37:51
this is true, the end to end immutability picture. And that's kind of why it's so simple yet, you know, so advanced under under the covers. And so often we talk about data protection and, and we think back up, we think the end of uh of the line, right? And I always like to, you know, we talk about it, backups and immutability are,
38:12
are fantastic and restore, restoring is even more important, but that's really your break glass in case of emergency, right? Like that's the last place. It's really the last place I wanna go mostly because it's the last place I have to go. Right. I'm I I'm not getting anywhere else after that. So that's, that's my break glass. I'm gonna go grab a snapshot,
38:33
a safe mode protected, snapshot, restore it. Beam can read it like, like Mark said, orchestrate it, read it because of their portable file format and I can restore from it. But we really need to start with this stuff on pro in production. Right? Let's make sure that our production arrays are protected and that they have a mutability if
38:50
I'm hit, the first thing I wanna do is restore from my production array, right? If, if I can and I'm missing some pieces, I wanna be able to bring those snapshots back immediately, right? And um, you know, Chris could probably tell you this, multiple customers, tell you this, the snapshots on, on pure storage are so fantastic.
39:09
They're immediate, right? There's no overhead, even when you take them, there's no space consumed, right? Because of our, and the only space that gets consumed is is change rate, right? As you're changing, you take that up, but restoring them is immediate as well. So that's where you want to start.
39:23
And I, I think that's really the, the, the picture here is we can help give an end immutability beginning to end like it says, and have all those great capabilities. Uh the, the just the way that we do safe mode and the way that VM has always done their portable backup file format and knows all about their files means you can have any of that data anywhere and beam and peer is gonna be able to read it and restore it for you.
39:48
Yeah. No great point. And 11 thing that came to mind, you know, we talk about ransomware a lot, which it's important, it's relevant but that's not the only threat, mistakes by administrators, not you, Chris, uh, but mistakes by those guys. Right. The other administrators, uh, it's,
40:12
it's a real issue. It doesn't happen a lot. I don't hear it as a, a common problem but it can happen. People make mistakes, especially, I've never wiped out all our domain controllers. Never. Yeah, we won't talk about what I've done. So, uh, yeah, it, it just, it's that additional, uh, you know,
40:33
it's belt and suspenders if you will, right? You, you can be pretty sure your pants aren't gonna fall down if you're wearing both. Anything else? No, I think with that, uh, I think we're pretty good. Yeah. So one thing, you know, you've got, uh, a flash array booth, you've got a flash blade booth.
40:55
So, you know, please go talk to those folks, uh, beam as you walk in the zone. We're over in the right side in the back. We've got some of our systems engineers here, technical experts in that are well versed and pure. So please leverage the folks that we brought if you have any questions, even if you were wondering if you wanted to talk to.
41:20
Who should you talk to? We can certainly help you with that as well.
