00:06
How's everyone doing out there? Come on, we could do better than that. How's everyone? Awesome, great to see all of you. So, uh, thank you so much, Charlie, for showcasing the vision behind the, uh, intelligent, uh, sorry, Enterprise data cloud.
00:24
I'm gonna be showcasing a little bit of the intelligent control plane. Pure has always been maniacally focused on simplicity since day one, and today we're taking that to an entirely new level. Our intelligent control plane enables you to be able to transform your business in 4 fundamental ways.
00:43
The first one is around infrastructure. As Charlie mentioned, we go from siloed fragmented environments to a virtualized cloud of data. The second fundamental transformation is around operations. Instead of you having to remember all of the manual tasks in order to deploy a workload, you can have it all pre-built as one intelligent, policy-driven workflow.
01:06
And then 3 is compliance. You need data governance, but you don't have it today. You deploy workloads, you hope for the best. We're gonna give you the world where you get data intelligence, you get the ability to report and remediate.
01:19
And then the last one is all around productivity. You get the ability to instead of doing the heavy lift of analyzing your environment, you now get to a world where you can just ask via natural language. Give me some help. I need to figure out where my performance challenges are. But I'm sure many of you are thinking this
01:36
whole intelligent control plane thing, it's been tried to be tackled before. I mean, lots of companies have tried to do this kind of thing. There's been the observability platforms that allowed you to see everything, but you couldn't really do anything with the systems. You had to do them manually, all of them separately.
01:52
There were virtualization platforms that gave you the illusion of a unified data plane, but it caused lots of complexity on bottlenecks and then you have this world of automation platforms which gave you the benefits of being able to reduce the manual steps, but at the end of the day it really was not really understanding what the back end infrastructure was. The honest truth here is that you actually need
02:16
all three of these capabilities in order to actually deliver an intelligent control plane. But it is still missing one key component out there, and that component is intelligence. Intelligence allows you to do a bunch of cool capabilities within the platform. You get the ability to be able to make better decisions on workloads where workloads are placed. You get to understand via data on what is going
02:41
on within the environment even allows you to be able to do things dynamically or in an autonomous way. The good news is that over the last 13 years and tens of thousands of systems, we've been consolidating some of the largest repositories of telemetry data out there and we've been using it for intelligence within our own environment. We build better products.
03:04
We build a support experience where you, you know, we solve issues before they even come up and we've been building a unique visibility into tier one and giving you the ability to be able to see your environment and understand what's going on. What we're doing now is we're taking that exact same intelligence and fusing it together with fusion to build our intelligent control plane.
03:28
Our intelligent control plane is architecturally unique. It's embedded directly into purity, allowing our systems to actually intercommunicate together as one, you know, meshed network of systems. They can discover, they can manage each other, they can actually intercommunicate to allow for dynamics that you could never do in a siloed fragmented environment in the past.
03:51
This gives you benefits, all sorts. You can first start to federate and build a cloud-like experience where you build that virtual cloud of data. Unex can go and you know build remote management capabilities or systems can manage each other. You can maximize the utilization of all of your
04:08
systems to get the best overall TCO and in the future will enable capabilities of being able to dynamically rebalance systems on the fly. And what's even better is we are very excited to announce that Fusion is available for Flash ray, Flash Blade, and Cloud Block Store today. So you guys can all, you know, do a firmware upgrade and get access to this and start
04:33
getting the advantages that I'm talking about. Alright, so now we talked a little bit about the infrastructure challenges, but let's talk about operations for a bit. I mean, I'm sure a lot of you out there, maybe after a late night in Vegas, someone calls you up and says, I'm gonna deploy an Oracle workload,
04:50
and you're like, oh man. What's the latency? What's the performance? What's the capacity requirements? Oh, it doesn't stop there, guys. What's the replication strategy? How often am I gonna, you know, hold snapshots and retain them?
05:02
What's my QS policy? What am I gonna tag it? I mean, the, the challenges are endless, but imagine a very different world. Imagine a world where you could have everything preset. All bundled together as one consistent way of deploying applications and make it so that you can actually deliver these in a highly consistent way,
05:23
but even better do so with intelligence. Pure fusion solves this challenge with presets, and these are also available for you to be able to leverage to give you day one compliance around consistency, leverage the intelligence within our systems to be able to deploy workloads not only consistently, but on the right system in the environment, and then provide in the future
05:48
capabilities of giving self-service to end users to actually be able to deploy these themselves. Sounds pretty cool, huh? You guys, you guys interested in seeing it? All right, let's check it out. I'd like to bring Matt up to the stage and showcase some of this cool technology. Matt, what's up?
06:06
How are you? Great. How are you doing? All right, let's jump in. All right, so look, we know a lot of you are still deploying workloads manually, and that's gonna come with things like chasing down performance, protection, compliance across dozens or even hundreds of arrays.
06:23
Now pure fusion on the other hand, uses real-time intelligence to automate precise policy-based deployments that are optimized for scale uh for performance and efficiency at scale. So let me show you how this works. Fusion turns your business policies into reusable blueprints called presets, and this defines exactly how the workloads are deployed securely and consistently.
06:48
So we'll deploy a mission critical workload under a preset using our standard naming convention that's the workload type, the storage class, and the protection level. And protection is baked in right from day one. In this case, we're configuring daily snapshots with a 14 day retention period. And if required, we could also include things like replication.
07:11
Next, we're gonna define the workload type so fusion can make intelligent placement decisions and you're gonna see in just a moment how that works. So we'll go ahead and choose Oracle and we'll move on to our storage class. Now think of this as a policy that defines where the workload is supposed to run, like only run on our high performance storage.
07:32
So we're gonna select the flash ray X and then move on to the storage configuration, and this is just where we can name and define our volumes. Like in this case we're creating 22 terabyte volumes and we'll apply that snapshot policy that we just created. And if we wanted, we could also add quality of service and tagging for things like
07:51
organization or even chargeback. And with that, that's the preset it's complete. It's pretty amazing that you can take all of these configurations and make it just, you know, stamped out every single time. It's great. Exactly. It's, it's consistency at scale. It's, it's exactly what, exactly what we need.
08:08
So deploying that workload is simple, right? We just use the new preset, give it a name using our convention, and fusion handles all of the rest. So here Fusion has already automatically chosen the best placement for this workload. Evaluating every array in the fleet that meets our storage class and using real-time telemetry from peer one to find the optimal fit that's based on performance,
08:33
capacity, and policy alignment. So let's deploy After just a few moments, our workload is live, even though it's on a different array. We have our two volumes fully provisioned exactly as the preset defined, and it's already in a protection group with snapshot and replication policies applied.
08:53
And because that group is protected by safe mode, those policies and the snapshots associated with it cannot be deleted even by an admin. So in just a few clicks, we've defined the workload, locked in protection, and Fusion deployed it exactly where it belongs. But you wanna know what the best part is?
09:13
You don't have to wait. Fusion is baked into purity ready for you right now. Very cool. I love that you can just auto deploy this on the right system even though you're not actually on that system and it lands it in the place that it should be, right? So this amazing technology works great within
09:29
the pure storage ecosystem, but what if it actually worked outside the ecosystem? Well, we're building in a single API within the system that allows them to be able to integrate and provide new values, leveraging the same technologies that we just talked about. In fact, one of the first ones will be VMware. You can go create a VMFS data store and leverage presets. It will auto deployed in the right uh system.
09:53
It will do it via a preset, and it's ensured to be compliant from day one. We're also super excited to talk about our new integration with Rubric. Rubric integrates with the Enterprise data cloud first by identifying indicators of compromise in their system. They then tag our snapshots with ones that potentially are suspect,
10:15
allowing users then to define which one's the best to recover from. We then build workflows to provide seamless recovery, and at the end of the day, this becomes a 1 + 1 equals 3 types of solution. You're gonna see a bunch of new technology partners that integrate into our platform and provide new values and services that enable you guys to be able to take your technology and
10:37
just make it even better with Pure. We're also thrilled to announce our new workflow orchestrator. Our workflow orchestrator allows you to take the same core capabilities of fusion, but now extend it out into thousands of different APIs into different ways of deploying applications. So in this case, let's say you wanted to deploy
10:58
a SQL environment, maybe have it replicated locally, maybe even have it replicated to the cloud. Matt, what do you think? Should we do it? Yeah, totally. I mean, if you thought fusion presets were cool, just wait, because let's kick it up a notch.
11:12
Now with fusion and Pure one forming the automation backbone, we can extend our automation capabilities and our intelligence beyond just storage all the way up to full application workflows. So let's walk through exactly what you described. We're going to start with a ServiceNow incident. We'll deploy a SQL environment,
11:28
replicate it locally, and we're going to send a disaster recovery copy to the cloud, all fully automated. So in our service now ticket we include the preset and a few key details about the request, and that's all that it takes. Our orchestration engine picks it up and kicks off the work flow immediately, and this process could also include multiple approval gates
11:50
giving organizations the governance that they need. So we hit save. Our request is submitted. Now while that runs in the background, let's take a look inside of Pier one and see what's happening. So here is the orchestration recipe which is made up of a series of intelligent connected steps, the first of which is the ServiceNow
12:07
incident which kicks off the workflow. Then we leverage Fusion's API. First we retrieve the preset and again this defines how the workload gets deployed. Then we pull a placement recommendation to ensure the workload lands on the best fit array. And finally, fusion provisions the storage volumes automatically following the preset and
12:28
the placement plan. And then of course we can send notifications via Slack and ServiceNow keeping everybody up to date and making progress tracking easy. Next, of course, we're gonna deploy the SQL database on one of our SQL servers, completing the core application setup, but wait, we're not done because since this is a gold tier workload, we're gonna add disaster recovery to AWS using Pure Protect.
12:53
So with that running, the fusion in the fusion fleet view, we can see the workload, the provision volumes, and the protection groups that are keeping it all safe and in Slack we can confirm success with a message showing that the database was created and is ready. So let's go check that out.
13:11
In SQL Server Management Studio, we can connect to the environment and everything looks just like we defined in the orchestration recipe. The database is in place, correctly configured and ready to store our data. So now let's move to Pure Protect and take a quick look at the applied policies. Here we're replicating one copy of the SQL virtual machine from our local VMware and
13:36
saving a copy in AWS EC2 every day and retaining that for 3 days for disaster recovery. And so just like that we've deployed a SQL environment locally and protected it in the cloud. So from request to protection, every step is connected, automated and built to scale to your cloud strategy. It's just the perfect example of how the enterprise data cloud delivers intelligent full
14:02
stack decision making just beyond just storage, so. Happy to say these capabilities are available in early access. Go check it out. Awesome, thank you so much, Matt. Thanks, Jeff. All right, so, uh.
14:20
Full stack application to infrastructure fusion natively integrated doesn't get much better than that. All right, so now we've tackled the infrastructure and the operations challenges. Let's dive in a little bit on how we handle compliance. So we've got this concept of telemetry from systems. We now have presets would tell you what the day
14:40
one compliance was supposed to be, but now we need to figure out like what is this thing actually doing in production? And so we're excited to showcase a little bit of some of the future of what we're going to build out for compliance. The first is a data intelligence engine. This will take data from your existing environment, understand it in much more depth,
14:59
and be able to provide things like application insights, what application it's running, showcase that across the entire fleet, create a data catalog so we have a semantics layer to understand exactly what's going on out there. Understand if you have sensitive data or even down the down the way look at ransomware detection. We'll be putting this into a new data
15:18
intelligence engine inside of Pier one that gives you full fleet views of all of this new intelligence and allows you to operate in a very different mechanism than just it's a block volume, you'll actually know exactly what's in it. Pairing with this, we're excited about the compliance reporting engine that we'll be building. Now you'll be able to go and define policies
15:39
that your company has, see where you stand on it, and be able to remediate it if needed. Just think about it for all the CIOs in the room, you have real-time auditing to figure out where you stand in the environment compared to where you expect the environment to look like. All right. So the final step here is productivity. We talk a lot about AI, but let's talk about how we use it.
16:00
We built this amazing technology called Co-pilot that allows you to be able to ask questions via natural language. Well, we've taken this to an entirely new level once again with adding in the telemetry data. So now you can go ask what would take in any cases, hours and hours to figure out and get answers in in mere seconds. Like, what's my busiest volume across 100
16:21
systems? It would take you forever to figure out that. I'm going to show you a demo of something that's even more fun here. So imagine you have a backup admin call you and say, Hey, we missed our backup window last night. Can you help me figure out where the problem is? Well, think about this for a sec.
16:37
You've got 100 arrays. It's gonna take you a couple of hours to figure out where the actual problem is, but instead you can just give it context. So let's go ahead and do it. So I'm gonna jump in here and say, one of my critical backup jobs missed his backup window last night. Can you analyze my fleet?
16:51
Find the performance bottleneck. It will go off and actually figure out where the bottleneck exists, give you a recommendation to optimize the system, and in fact, uh, you'll also see here the ability to be able to meet, uh, to fix this solution with uh increasing the performance.
17:09
So in this case maybe you increase the performance on something we'll be announcing later in the show, and it bumps up the performance of that particular volume to make sure that we can actually achieve the backup in time. It will ask for my approval because you want to make sure that these things are governed, and at that point, it will go off and make the change for you.
17:27
This is the future capabilities that we will be building into co-pilot that will allow you to be able to change the way you do storage management as a whole. All right, so we've talked a lot about the future of the enterprise data cloud and in this case the intelligent control plane. We've transformed infrastructure operations, compliance, and productivity with a ton of new groundbreaking innovations.
17:49
I want to showcase and have Rob Lee jump on the stage here next to showcase some of the innovations we're building into the unified data plan.
