00:00
Appreciate you all coming for, for our session here on some interesting things around automation that we're doing uh with Epic. Um, Chad Monti, the principal of the cyber resilience. TWs want you say hi to everybody and get going. Here, great awesome hi to everybody, let's get going, um.
00:20
So Tom Wheyland, um, I'm responsible for our integration of pure technology with the Epic, uh, use case as well as some other EMR technologies, uh, that we've we've worked on as well, um, but, uh, the focus of, of our time, you know, has been specifically around Epic as well as the inner systems Iris, and we're happy that the both customers of ours which is great.
00:43
um, I started out, uh, as a director of IT for a healthcare institution in Moss, Wisconsin, which is where I'm from. Um, and then, uh, I was then went to EMC for some time and then joined Pure in 2017, so, um, appreciate you guys all coming here and let's, uh, we'll kick this off. All right, so we're gonna talk about peer
01:06
Protect. For those that don't know, it's a kind of a new software offering that we have, but it does a lot of cool stuff within, um, just within the storage infrastructure to help you guys get better backups and let's see here, there's Chad. Did you, you introduced yourself, right?
01:21
OK, cool. So anyway, epic. Big red letters, um, so you know just to give you kind of a refresher for those that may not work with Epic, um, you know, it's a broad application it's a, I kind of refer to it always as a framework, um, and so because of that there's some inherent challenges and some architectural decisions that you really have to kind of think about but
01:46
ideally when we talk to customers about, you know, how to operate within this healthier space whether it's Epic or even other manufacturers. You know, these, all of these various points are, you know, really relevant to finding the right solution for those customers and, and I apologize I'm gonna go a little fast because we're kind of really behind time here.
02:05
But um the idea here is that all of the various kind of challenges that a lot of our healthcare customers face um we have some pretty solid answers to things that we're working on things that are currently being offered and a lot of lessons learned uh within our manufacturers as well as our customers so it's a it's a, it's a application use case that that we put a lot of effort into and uh our customers are are uh our most important assets so we're uh
02:33
offering that. OK, uh, I'm not gonna go through this one because everyone knows all about that. I, I'll cover this one, just a little bit here. What, what I'd like, uh, you know, everybody to, to take away from here is, you know, right now as we're managing these epic environments, you know,
02:48
there's a set of tools that have been in place for a long time. Usually they're script driven they're utilizing. Technologies on the array. They're utilizing uh command line interfaces to things like the database, the freeze, the the thaws, the business continuance with active cluster, maybe DR copies with active DRA sync replication, maybe taking clones of copies for
03:08
your reporting database or for uh development environments. These all are using tools that are very mature that are delivered by SLA within our arrays. So think of this as array technologies, active cluster, active DR continuous, snapshots, the first line of defense in any ransomware attack, the safe mode capabilities, async replication. Now we're adding Pure Protect as an
03:30
orchestration and automation layer on top of that. What we're gonna show you here, uh, is the change in these workflows away from the traditional manual process that's scripted. Or maybe that's driven by UIs or CLA II commands that you'll run, uh, that you'll run manually and how we can turn this into automation or essentially,
03:51
you know, epic operations as code if you will. So let's look at how we're gonna get there, um, Thomas back to you. OK, so one thing that's really important to understand, and this is a really busy document, but the idea, the takeaway here is that for a normal healthcare environment, whether it's epic or, or any other healthcare environment for that matter,
04:14
there's always a lot of moving parts and the moving parts are critical to make sure that they're all kind of be done optimally, otherwise, it makes the challenges of doing things like traditional backups, replication. Uh, and other kind of data movements, things like snapshotting and replication, all those data movement type of actions as you can see here,
04:34
there's basically 4 types of data movements being depicted here, but they all have their reasons for existing. Some of them are based on, uh, rules that every kind of ask the customers to be able to do things like using database level mirroring versus traditional kind of sand-based mirroring. Uh, as well as, uh, doing things like snapshotting, daily snapshots,
04:55
hourly snapshots, you know, customers all have their ideas about the best way to do that, whether it's, uh, you know, daily or hourly, it requires a lot of work to kind of script all that out and orchestrate it beyond simply just kind of using the tools that are delivered by, you know, your, your various storage manufacturers. It's all about how do you bring those tools together in a way that allows you to cohesively
05:17
kinda keep all of this stuff moving in the right direction. While at the same time keeping it all up and running and avoiding things like shifting failures, things like that so this is just is a I, I hope that this demonstrates the complexity that uh health care and specifically Epic customers run into uh when they're dealing with their their, you know, environments as they organically grow we saw this with COVID really
05:41
hit. It you know, all of these data growth type things that we look at all changed exponentially higher and so while COVID was a bad time, it was really hard for IT shops to really absorb the influx of additional uh user capacity that was demanded based on that situation. So imagine this, but kind of more to the extreme is kind of what everyone had to to go
06:04
through. So understanding that helps us understand, you know, where the problems lie and where, where there's opportunities potentially for some automational uh benefits that we can provide. So automation tomorrow and for today. So One of the things I talked to customers most about epic environments.
06:24
After they buy an array or hopefully before as well, is that scripting is a necessary evil. So you have to do it to some degree. Uh, I think is a great company, but they don't provide everything. So there is a there is a need to have some type of scripting, uh, processes in your environment in order to make your job easier for things like various
06:47
imports, uh, downloads from subscription providers, things like that to just fill in the data gaps that requires from those same companies. So, but a lot of the, the challenges a lot of customers don't have the kind of technical bench to be able to do that effectively. And have all that kind of intellectual property reside with one or two people because then when
07:08
they leave, the scripts still have to keep running. So we think that there might be a better way to do it and we're gonna hopefully show you some of that. Right now So the crux of this is showing a demo uh built within Pier one, utilizing our uh workflow automation and our Pureect tools to show that we can innovate in these areas without having to uh write and maintain scripts.
07:33
So we start this workflow job um within Pier one here and one of the first things we'll do and how you can drive this. It can be with something as simple as a Service now ticket. You can, uh, a user could go in there wanting to refresh their development database, executed a ServiceNow ticket, drive these workloads.
07:53
So in this case, we kick off the EPI workflow, um, and you see the there's a trigger. Now that trigger can be serviced now, it can be some other event that wants to go execute this operation. The next step that we just showed you is freezing of the database, creation of the protection group and the snapshot, and then the thaw of the database which we're gonna highlight right here.
08:16
Once that's done, whether you've taken that to go take a snapshot and refresh a database or do something else, we can post a message to a user in Slack that says what we've done. Here we're showing the flash array UI so you can see in the back end that we're actually executing these and successfully running them. Now all of this is managed where we can actually see the triggers,
08:36
the operations, and then monitor them. So you can take that output and you can validate it. You can put it into a tool, you can take action. If there's an error or some sort of other event, you can take an action on it based upon that. Uh, so in this we can actually show that we're grabbing and enumerating the cache database.
08:54
Taking the snapshot, monitoring it, while executing the freeze and thaw scripts as post and pre-processes. And then ultimately post the message to Slack saying that we've been successful. All of the stuff being represented and visible within uh Pier one as well, visible within the uh um the UI and then the end point backup successful protection
09:20
successful refresh of a database successful. Very quick highlight to show what we're building with our orchestration layer with the puretect pieces to say now instead of thinking about managing and operating, refreshing these environments uh with scripts that we have to maintain, now we can use a modern workflow automation engine.
09:38
To perform the same functions, have it easy to maintain, have it easy to monitor, and have it be able to execute all of those operations on demand. Service now kick it off, notify the slack, continue on. That's the crux of what we were wanting to to show everybody here today, um. You know, we'd like to hear from the room.
09:56
You know, what do you think about this, uh, this new approach to business continuance, disaster recovery, and epic? Is it an idea whose time has come, um, or do we like scripts? So one of the things we're also, you know, hoping to be able to do is to bring some customer involvement into the process.
10:16
So you know it's one thing to kind of go through uh uh a scripted demo um but it's another thing to have you know a real customer actually really using the tools and the capabilities and some of the functionality so if this is something that is interesting to you guys in your environments, um, whether it's epic or really any other data protection type of use case that you may have, we'd love to chat with you about that to determine if if
10:39
there's something we can do to help you guys kind of see the benefits of this at the same time. Give us feedback on, you know what the tool looks like what you would like to see in it that maybe it's not there today so that it it continues to to to inform our developers on the right direction for the product moving forward um because we listen to our customers um.
11:01
So any questions? Oh you know that was super fast, guys. I apologize we got kind of a late start, um, but, uh, if there's any questions, we're happy to answer them, um, or if you're looking for other information, Chad or myself, uh, I think there's a way that you'll have our contact information.
11:18
We'd love to hear from you, you know, offline, um, or get, uh, within the conference, so don't be shy. We're happy to answer any questions you guys might have. I think everybody up here knows how to get a hold of us. Um, we're looking for for yourselves in this room to engage with us, test this out, give us feedback, um,
11:35
let us know if it's gonna help you, uh, innovate and be a little bit faster, more automated within your work flows, and we can even use this for a lot of other interesting use cases. So thank you everybody. I know we're running probably about 5 minutes late here still, but we're gonna stick around for a little bit, um, and let us know what you think.
11:51
We'd appreciate it. Thanks.
