00:00
Hi, I'm Zane Allen, principal technologist with Pure Storage. And I'm here today to talk to you about your and my favorite subject, data protection. I know. I know there are plenty of other things you'd like to be thinking about and talking about, but you did click the link.
00:16
That's probably because you also know that as an IT professional, it is your responsibility to keep your organization's data integrity intact. But dealing with everything you have to deal with today, like sprawling environments, expanding, changing virtualized workloads, whether cloud fits in your environment or not, all while dealing with growing threats,
00:36
it's everything you can do to keep your organization's SLAs intact. And it could be a real challenge to find the right DR solution. And just stitching together point products just doesn't cut it. You need a protection strategy that is resilient, efficient, and as integrated as those environments that you're protecting.
00:53
And this is where VMware and Pure Storage on VMware come not as standalone tools but as a cohesive platform for the modern data center. So today I'll show you how these technologies do exactly that, and more importantly how the VMware and Pure integration uniquely unlocks the level of automation, performance, and reliability you simply can't get without the sum of those parts.
01:20
So, we start with the old debate: our snapshots backup, right? The whole question of which backup strategy is best. If I have storage that has snapshots, can I not just use them? If I have a backup data protection that's outside of storage, can I not just use that for backup and restore?
01:37
Can I not just use the built-in data protection tools my hypervisor uses? I mean, it's included, right? And the answer is no. And I guess it would be odd if it weren't. And why that is is though those tools on their own can offer the resiliency you may need in
01:52
some cases and the performance you may need in others, they can't quite cover all the restore scenarios entirely on their own. And so what do I mean by that? Well, first, let's take a standard VM backup or a snapshot, if you will. And as you all know, snapshots and hypervisors they just leverage a file-based solution,
02:11
right? These are files, with logs, things being written to a data store, to a file system, be it VMFS or NFS or whatever it may be. So when you take a snapshot inside VMware, either temporarily backup or some long-term recovery strategy. I don't know, you're doing an OS upgrade on a VM, let's say.
02:34
What you are doing is you're writing an external or an extra amount of data that now has to be dealt with. Now it has to be consolidated at some point. And while it's open for someone to access that production VM, that file, along with the base VM file, has to be read at the same time,
02:51
and it causes a lot of IO and problems for load on the storage, potentially causing interruption of service. To the application, translation, trouble tickets, right? Call to the help desk, and it could take a long time to clean these things up, depending on how long they're open.
03:05
So integrating these, a VMware snapshot that is with your storage, either for backup operation or just some good data protection, cuts down that process considerably, making the actions needed for good data protection a negligible process. I could temporarily create a file for data integrity capture for backup,
03:24
and then quickly erase it. This removes the performance strain on the storage, ensuring efficient operations and resilient data protection, right? I can capture that data in the state I want to without putting that load on production. So it's great, but why can't I just use snapshots on my storage?
03:40
My storage has those built in, right? That's outside of the VMware environment. I don't have any of these redo logs, these consolidation processes, all those limitations you talk of, saying. And with pure snapshots, I know those are instantaneous. This is just a metadata operation.
03:56
It uses a very limited amount of space. But remember, a snapshot of a volume is the entire data set. Not to mention that on their own, they do nothing about getting the application into a current consistent state, right? It's only what we call crash consistent. And that may not be optimal for your line of business applications.
04:16
And even if it were, when you do a restore. You have to do a complete restore or you have to figure out some manual operation to pull the data out that you need, or you have to clone and restore to a new instance, right? Just adding further time to the restore process, which, when disaster strikes, you may not have the luxury of more time in order to get your
04:34
business back up and running. Now the ability to recover quickly in bulk is exactly what might be warranted at times, but it doesn't fit every situation because there isn't just one recovery scenario out there. There are many depending on what's happening in the environment, be it a cyber event, a complete site failure,
04:53
whatever it may be. So what you need is something that is integrated to deliver on that. OK, so how do I just restore what I need then, Zane? Do I just go to my backup file source for my Snap? Do I have to account for the extra space to keep copies everywhere?
05:08
I may need to restore, and as you're a storage company, you're happy to sell me more gigabytes to cover? How does this integration really help me with this? I'm glad you asked. What if I told you you could use both? What if you could restore the pieces you need from the snapshots,
05:25
snapshots that are close to the source, right, get that data back quickly, as well as have the option to dig into the longer term backup files to restore from an archive, maybe adding a different site or even convert to another format for any reason you may need. Sounds good, right? Well, then let's get into how these pieces work together to provide a recovery solution for
05:46
every type of file restore scenario. First, we'll break down the components because it is the architecture that makes the difference. We have VMware sitting on top of Pure Storage at that hypervisor layer that is using Pure Storage, most likely block, but it also could be done with NFS.
06:05
What we then do is tie V into this mix to essentially stitch those integrations together, and there are a lot of integrations Pure has with VMware today and continue to develop on, as you can see demonstrated in other videos. Um, and now configuring VAM in this manner. In allowing them to understand that integration, that relationship,
06:28
how those two are put together, what components are available across the infrastructure, the overall architecture, if you will. For instance, it can now understand that the snapshots that we talked about underneath on the storage layer, it can make sure that they are enabled with safe mode, that they aren't deleted before they're required, and it can start to use those
06:48
snapshots for things like those backup operations, as well as for restore. I can pull data back very quickly. I can back up data on a schedule that makes sense for my organization, and I can do all of it without impacting any of the production environment. I can keep snapshots for a longer-term kind of close to the data,
07:07
having VAM orchestrate the movement of all of it. But what about the needs that may require extending the exact same data recovery scenarios into my DR site? Well, once VAM understands that relationship. It's orchestrating some of it. It starts reaching in and indexing those
07:26
snapshots, understanding how the vSphere and the Pure environment are working together. Now it's looking deeper into the organization, wider into the organization to provide advanced DR capabilities for those outages that might require resources outside of the production environment. With VAM integrated, I can extend all the features mentioned above.
07:47
The will ask here to take snapshots from the production array. It'll send it, and then it can send it to the secondary site, and then I have all those same capabilities. I can keep them there for instant recovery at DR, use them as the source for a backup at DR, and keep that backup for longer term. All in this solution that's stitched together very easily,
08:07
extending the recovery well beyond production. With this solution, you're provided the ability to restore for many different DR scenarios ransomware events, full outages of sites, just lost data, deleted data, insider threats, and you have all those recovery points that you can perform more frequently with snapshots, and they can do more periodically,
08:29
maybe say daily with long-term retention with the backup files, all while not impacting production. If this sounds good, don't you check out some of our next videos in this series and learn a a little deeper into how this works. Thank you for watching.
