Latest / Nutanix Weekly / Nutanix Weekly: Storage Mobility to Meet the Challenges of Hybrid Multicloud

Transcript

Nutanix Weekly: Storage Mobility to Meet the Challenges of Hybrid Multicloud · Nutanix Weekly

WEBVTT 1 00:00:03.300 --> 00:00:13.200 Philip Sellers: Hello, and welcome to another episode of Nutanix Weekly. I am your host, Bill Sellers, the practice director for Modern Data Center here at Zintegra. 2 00:00:13.627 --> 00:00:21.960 Philip Sellers: Happy to have you joining us for another episode, and another discussion about Nutanix and the ways that it can help. Your business succeed. 3 00:00:22.314 --> 00:00:41.189 Philip Sellers: This episode is one of 12. I think we have now at Zintegra. We like to call it content with context. And that's because we try to take content from other places like the Nutanix blog that we'll be covering today, and then add the real world to it. Add our own experiences and our insights to it. So 4 00:00:41.340 --> 00:00:48.169 Philip Sellers: I've got 2 great nutanix resources on the line with us today. Mr. Jaira Cox, how are you doing. 5 00:00:48.570 --> 00:00:50.200 Jirah Cox: Bill. Doing. Well. Man. Happy Monday. 6 00:00:50.880 --> 00:00:55.069 Philip Sellers: Happy Monday, and we've also got Ben Rogers. Ben, how you doing, son? 7 00:00:55.400 --> 00:00:58.649 Ben Rogers: Hey! How are you doing? Long time to see guys? Glad to be back. 8 00:00:59.210 --> 00:01:00.679 Philip Sellers: Yeah, it's good to see you. 9 00:01:01.067 --> 00:01:10.979 Philip Sellers: So Jaira and Ben both work for the company, both work for Nutanix. So it's great to have you guys with us today. We've had 10 00:01:11.170 --> 00:01:15.999 Philip Sellers: a lot going on in the Carolinas for anyone listening. 11 00:01:16.429 --> 00:01:23.550 Philip Sellers: We we faced tropical storm hurricane, Helene. I'm not sure what it was by the time it actually got to us, but 12 00:01:23.940 --> 00:01:25.929 Philip Sellers: we saw the effects of it 13 00:01:25.940 --> 00:01:35.699 Philip Sellers: a lot sooner than the actual. I came through our area. We've got a lot of people across the country that have been affected by that storm. So 14 00:01:36.127 --> 00:01:39.789 Philip Sellers: how did you guys fare where you're at. 15 00:01:44.970 --> 00:02:00.899 Jirah Cox: Good thankfully, very grateful to say no effect. Over here on the Raleigh area. We had a rainy, windy day, but no totally lots of sympathy for folks that are having much more interesting days than we're having today. Hope you guys get through that quickly. And and I know my support's already been sent over to you toward y'all. 16 00:02:02.380 --> 00:02:10.979 Philip Sellers: Yeah, we. We've got team members here at integra that live up in the mountains of North Carolina. Have been participating in 17 00:02:11.280 --> 00:02:22.662 Philip Sellers: rescue efforts and things like that. There's a lot of devastation all throughout the eastern southeastern area. So thoughts and prayers with everybody affected. 18 00:02:24.270 --> 00:02:30.148 Philip Sellers: But we're we're here to talk about storage. We're here to talk about nuthanix 19 00:02:31.010 --> 00:02:37.409 Philip Sellers: So let's let's talk about today's blog post storage mobility to meet the challenges of hybrid multi cloud 20 00:02:37.824 --> 00:02:41.810 Philip Sellers: hybrid multi-cloud is something, I think, that we all agree is here to stay. 21 00:02:43.260 --> 00:02:45.410 Philip Sellers: companies can't get around it 22 00:02:45.630 --> 00:02:52.789 Philip Sellers: better or worse. We we have hybrid multi cloud. Right? We're we're doing compute in multiple places. 23 00:02:53.000 --> 00:02:56.430 Philip Sellers: whether that's on Prem manage service providers. 24 00:02:57.180 --> 00:03:10.949 Philip Sellers: azure. Aws, we've got multiple places that we're trying to get the job done trying to get work done. So this blog post was written by Dwayne Lesner, so shout out to Dwayne, principal technical marketing engineer for Nutanix. 25 00:03:11.190 --> 00:03:18.520 Philip Sellers: And I guess let's let's dive in Jared, you brought this one to to the table. Do you want to kick us off with it? 26 00:03:18.870 --> 00:03:38.219 Jirah Cox: Totally. And of course, thanks, as always, to our amazing tech marketing team, we have really a embarrassment of riches, right of great blog posts to choose from right now for what we want to cover. So, thanks again to the whole team, we can definitely shout out specifically to D. Wayne, I think it's how he pronounces it. So thank you, sir. 27 00:03:38.220 --> 00:03:53.429 Jirah Cox: So today's blog post. Yeah. Talks about. Well, the title is storage mobility to meet challenges of the hybrid multi-cloud, I think. Yeah, I mean, put very simply, it's like, how do we help you move and adapt your personal hybrid multi-cloud 28 00:03:53.510 --> 00:04:12.410 Jirah Cox: with whatever shape you want it to be, and take whatever you want to, wherever you want to go, right, whether that's apps vms, we'll touch later on about like some specific euthanix functions like unified storage, which we we covered in in deep detail in our last 29 00:04:12.570 --> 00:04:32.829 Jirah Cox: episode. But in general, like, how do we help you keep the shape of your cloud, very cloudy and less rigid right, and adapt it when you, when you want to. The article, even even starts off by talking about, you know, just flexibility required right? As different things like data, growth, like acquisitions. 30 00:04:32.830 --> 00:04:50.629 Jirah Cox: even just inherent it. Complexity make things want to change right? And and you know. Of course, in some ways every problem can be solved with the right kind of credit card swipe, and we also know that you know you do that enough, and you get invited to reevaluate the way you're deploying technology. 31 00:04:51.530 --> 00:04:56.339 Philip Sellers: Yeah. I mean the simplicity of the credit card swipe right? That's easy. 32 00:04:56.420 --> 00:04:59.980 Philip Sellers: But it induces its own set of problems because 33 00:05:00.120 --> 00:05:08.346 Philip Sellers: those environments they spin up. They're not necessarily to standards. They they don't meet our security requirements. 34 00:05:09.640 --> 00:05:15.770 Philip Sellers: but they're easy, right? And I think that's that's the underlying thing. When we talk about Cloud, it's not a destination. 35 00:05:15.920 --> 00:05:18.080 Philip Sellers: It's a way of doing things. 36 00:05:18.100 --> 00:05:28.250 Philip Sellers: and that way of doing things is actually pretty simplified. It should be easy to consume. It should be easy as swiping that credit card. 37 00:05:28.470 --> 00:05:36.109 Philip Sellers: and that's the goal here with nutanix infrastructure is to give you that simplicity, deliver that simplicity 38 00:05:36.280 --> 00:05:40.050 Philip Sellers: of consumption to your customers. At the end of the day. 39 00:05:40.350 --> 00:05:44.589 Jirah Cox: And I think as an it leader, right? There's a mandate to get that right. Because 40 00:05:45.270 --> 00:06:11.109 Jirah Cox: you, we need to make sure that we need. We need to make it easy for our customers to do the right thing the 1st time, right? Because some things are are brittle enough, are stack enough, have enough weight and mass around them, not looking in the mirror. But I do, too, around like let's get it right the 1st time, because it's hard to go back and solve and change later. Right? So when we make it, we make doing the right thing, the easy thing to do, then more things get done right automatically, or or on their own power. 41 00:06:12.360 --> 00:06:21.429 Ben Rogers: So it's interesting. I had a conversation, this weekend and event here in Charlotte, and the company was basically telling me their cloud journey started 42 00:06:21.830 --> 00:06:46.689 Ben Rogers: by a developer not wanting to wait on infrastructure. And so he started doing cloud infrastructure did more and more and more, and the next thing they know they've got, you know, workload that's sitting in cloud that they're now responsible for securing. And that was the start of their cloud journey. Is this developer just got way ahead of himself before he knew it had multiple projects in the cloud, and they were playing catch up. 43 00:06:46.710 --> 00:07:10.350 Ben Rogers: And they're struggling to get, you know consistency between on Prem and cloud. And so it sounds like, you know, not reading too much of this because we're supposed to read it while we're going through it. It sounds like that. We're able to solve some of those consistency problems and give you a platform that will allow you to do both on Prem and cloud, if need to be, or migrate between the 2, if need to be. So I thought it was interesting that 44 00:07:10.450 --> 00:07:26.079 Ben Rogers: not all cloud, not all organizations have their clouds defined for them. It's usually sometimes a reaction to something that's happened in their environment. Like with this particular customer, they had no Cloud plan until their developer got tired of waiting on people. 45 00:07:27.380 --> 00:07:34.989 Philip Sellers: Think that's actually a pretty common sort of scenario, right? The barrier to entry is really low, and 46 00:07:35.040 --> 00:07:37.970 Philip Sellers: we get into the situations where 47 00:07:38.000 --> 00:07:45.109 Philip Sellers: we. We have traditionally taken too long to deliver resources to our constituents, you know, as infrastructure folks. 48 00:07:45.400 --> 00:07:50.319 Philip Sellers: and that is frustrating. They're trying to write code and get the job done at the end of the day. 49 00:07:51.700 --> 00:07:55.920 Philip Sellers: But this is where we also need the other pieces that are 50 00:07:55.960 --> 00:08:06.699 Philip Sellers: required of us, required by compliance required by business, by just good stewardship of the data and resources we're given by our business partners. 51 00:08:06.780 --> 00:08:13.929 Philip Sellers: We have to do the right thing around this and that may not be front of mind when you're just going out and and starting to consume things. 52 00:08:14.090 --> 00:08:18.229 Philip Sellers: But there's also the skills gap right? And and we'll talk more about 53 00:08:18.270 --> 00:08:42.340 Philip Sellers: all these other concerns. But there's that skills gap right? And I think you highlighted it. There been a lot of traditional infrastructure teams. They're not necessarily equipped to just inherit a cloud deployment and then know what to do. From a governance standpoint. Security standpoint. These things operate differently than our traditional infrastructure has or 54 00:08:42.679 --> 00:08:52.870 Philip Sellers: does it. And maybe that's where the solution is in all of this. So let's check on a little bit here in in the blog post and you know. 55 00:08:53.450 --> 00:09:12.080 Philip Sellers: we we go on to learn. And and I think we've talked about it many times here on the podcast but Nutanix cloud clusters at the end of the day. It's the same software that you run on Prem, and so let's talk a little bit more about nc, 2. As a solution for that consistent 56 00:09:12.360 --> 00:09:17.229 Philip Sellers: operating environment. No matter the destination. Jaira, I'll I'll throw it to you. 57 00:09:17.650 --> 00:09:33.350 Jirah Cox: Sure, so simplified diagram here at the top of the blog post covering. You know, kind of 3 corners of the triangle from on-prem to hyperscalers. Right? You know. Think about azure, think about aws right? Your favorite standard. 58 00:09:33.430 --> 00:10:02.710 Jirah Cox: you know, quote cloud, unquote, right? Like traditional cloud. What people think of when they hear about that. And of course, service providers right? And in reality, of course, with Nutanix with our software platform, all 3 of these things can become your cloud platform. It's a matter of you know. Do you want to, you know, have hardware. You own that lives in 4 walls that you own. Do you want it to be hardware? You own in 4 walls that you rent access to, or hardware? You rent in data centers that you also rent access to right all these. All these things are possible. Right? A matter of how do you want to get deployed? 59 00:10:02.740 --> 00:10:08.139 Jirah Cox: But how does using this ultra platform, then let you adapt and and evolve 60 00:10:08.320 --> 00:10:28.020 Jirah Cox: the underlying. Substrate, as you don't have to evolve and keep offering consistency to the business right? So I still vend vms, I still vend security, governance, optimization, visibility, right or observability. Responsible economics. As I can revisit any of these things right? Depending on 61 00:10:28.020 --> 00:10:42.067 Jirah Cox: economic factors, laws of physics right? I need things closer to other things. I need I mean, you know, not to be too put too fine a point on it. But when, like major freak storms happen right, I need to move workloads out of their path. 62 00:10:42.760 --> 00:10:48.120 Jirah Cox: in some ways that some of us here in the Carolinas are probably a little bit jealous of. I wish I could just like Wake up tomorrow and be across the country. 63 00:10:48.496 --> 00:11:03.700 Jirah Cox: You know. How do I offer that kind of mobility of the business with consistency. Of the platform that they're consuming right? Which, of course, can be the Nutanix platform itself. So that workload mobility is. And and I believe this Duane calls it out explicitly. 64 00:11:03.970 --> 00:11:24.169 Jirah Cox: is going to be a persistent requirement, right? The only constant is change, right. So there's nothing that we can trust is going to stay that way forever. How do we adapt and move, or we use the dumb analogy sometimes of, like, you know, rebuild the wings on the airplane as we fly it, or change all the tires on the bus as it rolls down the road. So what what can we do without causing disruption. 65 00:11:24.170 --> 00:11:40.690 Jirah Cox: as we actually are doing major work under the covers, to then replatform things, to move things, to reshape things. And so, of course, an easy one. Right is is using public cloud resources, right? Including, let's say, using bare metal nodes like call them just what they are servers 66 00:11:40.850 --> 00:12:06.759 Jirah Cox: from cloud providers with Nutanix as a Dr. Site. Right? I can have my workloads in one site replicate them to the other, and of course, when needed, just activate them, replicate them over, power them up, boom! I've accomplished my Dr. Objectives. The rest of this will walk through some some other ways that you want to do that a little bit differently, but at a base level, of course, that's 1 thing you can do is just flat out. Run Nutanix in public cloud as a Dr. Or even live data center placement. 67 00:12:08.270 --> 00:12:24.580 Philip Sellers: When I I think that's a great place for us to start is that data mobility part of the discussion. So we we're talking about Dr. Which uses native replication features inside of the Nutanix platform. That's not the only way that we can talk about data mobility. I mean, Ben. 68 00:12:24.770 --> 00:12:34.129 Philip Sellers: There, there's a great tool. We we talk about it a lot with customers, and that's Nutanix. Move. That's a huge advantage. A huge feather in the Nutanix Gap, too. 69 00:12:34.970 --> 00:12:42.160 Ben Rogers: Well, and that's a that's a Swiss army knife, because that not only takes you from Nutanix to Nutanix, but it can take you from several different 70 00:12:42.310 --> 00:13:01.849 Ben Rogers: competitors that we have into Nutanix. So whether you need to come from Esx in the Nutanix, or some of the cloud providers are into nutanix. It gives us a way to basically, where do you need to come from? And how can we absorb you into this? Another thing, I would like to say, is specifically to the diagram that was provided to us. Here 71 00:13:01.850 --> 00:13:24.779 Ben Rogers: is the different types of storage platforms that we have. It's just not volume and block. We can do object stores now, and you know we're handling unstructured data and giving security around that. So it might not be. You know you might want. You might run one data type in one data center and need to leverage another data type in cloud or 3rd party or disaster recovery. 72 00:13:24.830 --> 00:13:46.070 Ben Rogers: And you see here how he's been, you know, got one data center running one type of storage and then another data center. He's offloading into object stores. So again, it's not just the 1 1 size fits all we have the flexibility to be able to go. What what flavor do you need to see because of your operational restraints, and we have the ability to deliver that. 73 00:13:48.500 --> 00:14:07.209 Philip Sellers: Yeah. And it's a great call out right? This is a unified platform with lots of different ways to consume. But the one thing it's not. Is Hotel California, right? It's easy to get in. But it's hard to get out. Nutanix. It's easy to move that data around, and I think that's 1 of the key features we're talking about here 74 00:14:07.440 --> 00:14:13.240 Philip Sellers: with hybrid multi cloud is the data portability around all those different locations. So 75 00:14:13.665 --> 00:14:32.049 Philip Sellers: the blog calls it out a little bit further up. But you know, if you've got a cost advantage of one provider over another. Well, it's easy to move. Just move that data using native services and you get the advantage of that better pricing from Provider X 76 00:14:32.475 --> 00:14:45.239 Philip Sellers: or let's call it provider wide. So you, you get to consume on your terms, and you get the independence of being able to migrate things as needed. 77 00:14:45.870 --> 00:14:49.790 Philip Sellers: That's not the only sort of advantage. And 78 00:14:49.910 --> 00:14:52.320 Philip Sellers: actually, before I go there, you know. 79 00:14:52.650 --> 00:15:08.619 Philip Sellers: we're talking about move here. There's another great use case for Nutanix move, and that is customers trying to migrate away from Vmware Cloud on Aws into Nc. 2. It's a great pathway for you to move from a very similar sort of 80 00:15:08.840 --> 00:15:11.330 Philip Sellers: technology stack where. 81 00:15:11.830 --> 00:15:17.159 Philip Sellers: you know there's some difficulties going on right now in terms of licensing and 82 00:15:17.220 --> 00:15:20.359 Philip Sellers: relationships between broadcom and 83 00:15:21.044 --> 00:15:30.760 Philip Sellers: aws. So if you're looking for a good partnership, Nc, 2 on aws is is a great place to relocate those workloads as well. 84 00:15:30.780 --> 00:15:32.040 Philip Sellers: and one of the. 85 00:15:32.040 --> 00:15:35.770 Jirah Cox: I would say easiest ways to even try out Nutanix. Right? You're already running. 86 00:15:35.770 --> 00:15:36.159 Philip Sellers: Oh, yeah. 87 00:15:36.160 --> 00:15:41.199 Jirah Cox: Your workloads in a in a hardware and software as a service environment 88 00:15:41.880 --> 00:15:54.870 Jirah Cox: that that would be a complete like. For, like match of, I can go spin up a new cluster for a few more nodes paid for by the hour. Try that out, I think the nutanix part of that even is offered as a free trial. Right? So you're just paying for the cloud instances 89 00:15:55.100 --> 00:16:11.390 Jirah Cox: and move copies of your vms over without even disrupting them or turning them off if you wanted to right and test out what would look like to run this workload, this Vm. On Nutanix as well. One of the easiest ways to to take a test drive. It'd be akin to. 90 00:16:11.460 --> 00:16:21.260 Jirah Cox: I'm on vacation in a city. I can call up an Uber or a lyft and see that car that I think I might want to buy, you know, with 0 commitment and no warning instantly. 91 00:16:22.870 --> 00:16:24.190 Jirah Cox: Yeah, exactly. 92 00:16:24.190 --> 00:16:47.469 Ben Rogers: And this is not a rocket science man. I just did a demo of the migration plan. So I mean, basically, your source would be an Esx or an Aws, or azure, or something like that, and then your target would be the Nutanis cluster. What move does for you? Is it does all the virtual hardware mapping for you. So when you go to convert that thing from one format to another, we're taking care of all that for you. And then to Jayra's point. 93 00:16:47.470 --> 00:16:58.980 Ben Rogers: You can set these things in a test bubble. So you can actually create a migration plan. Let the Vm migrate over and keep the the original Vm. Running and test in a bubble 94 00:16:58.980 --> 00:17:20.570 Ben Rogers: the new Vm configuration, once it comes over, so move does a lot of cool things for you and takes care of a lot of the underlying infrastructure with the virtual hardware that you just don't have to worry about man. Just basically point it to the environment. It'll do the conversion for you. And then in the end, you know, people go. Okay. I've got the Vm. On the Esx side, and I got the Vm. Over on the Hv. Side. 95 00:17:20.569 --> 00:17:37.919 Ben Rogers: We have a cut off period, so we'll keep the data in sync for you until a given time, and then when you have a small outage window, we'll do a last sync and then bring that Vm. On the Hv. Side. We won't destroy the Vm. On the Esx side. We'll just down it for the period until you make sure that everything is successful. So 96 00:17:37.940 --> 00:17:51.159 Ben Rogers: move handles a lot of the migration tasks for the engineering and creates a good migration plan that can be repeatable, as you, you know, get successful. Some workloads and need to move other workloads over. 97 00:17:52.780 --> 00:18:01.949 Philip Sellers: When I think that follows the common story here, which is simplicity and orchestration, and that's 1 of the tenants, and and I try to talk about it often 98 00:18:02.480 --> 00:18:12.849 Philip Sellers: from day one day. 2 ongoing. That's 1 of the major underlying features of the Nutanix platform is that simplicity and orchestration? 99 00:18:12.980 --> 00:18:18.219 Philip Sellers: You know it's easy to add a node. It's easy to upgrade the nodes. It's easy to 100 00:18:18.782 --> 00:18:21.397 Philip Sellers: handle any one of these 101 00:18:22.110 --> 00:18:24.709 Philip Sellers: sort of day. 2 operations. 102 00:18:26.080 --> 00:18:38.649 Philip Sellers: which kind of brings us to another day. 2 operation. These bare metal nodes are not necessarily stuck with a finite configuration when it comes to storage anymore. Jaira, do you want to tell us a little bit more about that? 103 00:18:39.060 --> 00:18:40.219 Jirah Cox: Yeah. So 104 00:18:41.200 --> 00:19:05.489 Jirah Cox: the perhaps in the past, folks that have tried out our Nc 2 offering right have shown it's very easy to buy, you know. Hey? Do you want the small, the medium, the large right, the right off the rack? How many do you want to make your cluster? Well, what if you're like a sh medium right? And need something kind of in the middle there, right? How do I get something that's kind of like kind of right size for me when I just need more, let's say, like storage. So of course, these are these are, you know. 105 00:19:05.550 --> 00:19:35.249 Jirah Cox: regular servers in, you know who knows what data centers living somewhere in the cloud. There's literally no amount of money. You can pay a human to go and, like Pop more disks into them the way you would like. Of course, your nodes in your data center. So what's cool here is we actually use a similar principle to to solve for this kind of problem right? With like, what can we get as a service? Right? So of course, in Amazon, we can use ebs right elastic block storage that is also available in in cloud data centers to then present more additional block storage to that node. 106 00:19:35.290 --> 00:20:03.849 Jirah Cox: Again, all defined in software, all as an all as a service. So this is more akin to like. Think back to when we would have a node that might be like, say, Nvme hot tier with SSD colder quote quote tier SSD. Is still pretty darn fast right? But it is not quite as fast as Nvme on a on a you know, megabyte for megabyte basis. So now this node can have all the internal storage it already had before, can also have additional block storage. Also living in the Amazon data center as well 107 00:20:03.850 --> 00:20:32.580 Jirah Cox: presented as local storage. Cvm. Sees it all. Our Nutanix Controller, Vm. Sees it all, and aggregates all of that, just as if it was additional disks living inside of that node. And of course, our original bit of technical wizardry from 15 years ago right profiling vm workloads, placing the very hottest data on the very, most local, highest performing storage, letting other more randomly accessed data live elsewhere in the cluster in this case can be even on block storage 108 00:20:32.640 --> 00:20:39.410 Jirah Cox: in the cloud provider. There now we now we can bring more storage to bear than even what fits into the node itself. 109 00:20:39.470 --> 00:20:48.269 Jirah Cox: Right that lets our customers now fit even more economical. Dr. Use cases. Go back to our framing device of let's say you need to solve for Dr. And you've chosen to use Cloud to do it. 110 00:20:48.890 --> 00:20:54.019 Jirah Cox: Now, if I needed, say 10 nodes before to fit my storage capacity, I wanted out of my cluster. 111 00:20:54.140 --> 00:21:10.890 Jirah Cox: Maybe I can get by with 3 nodes with more storage attached to those things, and not need as much, not need as many nodes involved at the party just to hit a storage target when those nodes are mostly going to be asleep. Being Dr. Hosts right? They normally wouldn't be running vms and be very busy. 112 00:21:11.060 --> 00:21:23.529 Jirah Cox: so wonderful both, I think, I think, actually very cool, technical level of solution, but also huge financial punchline there on. Why would I care about it? Why would I? Why would I use it? What does it look like to run this kind of environment at scale? 113 00:21:24.320 --> 00:21:34.029 Ben Rogers: So at this point, crowd, we're hitting the awesomeness button, because that's awesome. Technology is being worked here. Add storage on the fly and not add clusters. 114 00:21:35.510 --> 00:21:58.797 Philip Sellers: Yeah. And and you know that that's the the other thing, too. Like as we talk about this, you know, we're talking about block storage. But that's only the interface coming into the host. So we can take that block storage, and then we can make it used as object and file on top of it, and that's the next section of the the blog post here, as we keep trucking in awesomeness. Just to quote Ben 115 00:21:59.190 --> 00:22:13.329 Philip Sellers: we we were talking about file services and scale out file services on Nc. 2. And you also have the ability to now do that on native Ec. 2 or Amazon elastic compute instances as well. So 116 00:22:13.857 --> 00:22:18.889 Philip Sellers: that's the next great feature is that you get file services. 117 00:22:18.910 --> 00:22:25.070 Philip Sellers: not even necessarily on the Nutanix platform, but on native cloud services, and it's your choice. 118 00:22:27.210 --> 00:22:36.550 Jirah Cox: 100%. Yeah. So one cool. I can think of one great use case for this. We actually with a customer. We all know that uses the cloud for for their Dr. Target for Vdi, right where those profiles 119 00:22:36.630 --> 00:23:04.090 Jirah Cox: you do your replicating all the time. But the actual hypervisor level running those desktops is going to be asleep most of the time outside of a test or a live failover environment, and so that ability to to pick a certain kind of workload like in this case, files operating as that. Smb nfs, unstructured data host can now run outside the cluster right? In a more economical, more effective way to deploy that and not have that be the workload. That sort of keeps my cluster larger than it needs to be most of the time 120 00:23:04.477 --> 00:23:14.299 Jirah Cox: is is super efficient for customers, right? So that that means I can now run just files. If I wanted to to have Cloud be my Dr. Target just for the files workload. 121 00:23:14.700 --> 00:23:39.189 Jirah Cox: or have that run next to a smaller cluster as well. If I needed, say, both like file replication for my profiles and vdisk replication for my gold master images, or persistent desktops or server vms or things of that nature. So yeah, it's a wonderful wonderful new feature with that ability to run files now natively on cloud itself. Your choice as a customer, either on Nutanix in cloud, or next to Nutanix in cloud. 122 00:23:40.240 --> 00:23:47.000 Ben Rogers: I think what this shows is just our company's commitment to to being a Cloud provider platform 123 00:23:47.030 --> 00:24:11.880 Ben Rogers: and being able to roll out individual services and not have to take the whole portfolio. And so this is where I really see our commitment to going. Okay. We know that cloud platforms are our portfolio of products and a portfolio of services. And here's our commitment to delivering the services the way our customers need to consume it, not the Nutanix way, but the customer way with simplicity 124 00:24:11.940 --> 00:24:25.439 Ben Rogers: and flexibility. And so that's really what I see. This whole files and and native cloud being is just our commitment to bring flexibility to our custom customers, to how they need to consume our data services. 125 00:24:27.250 --> 00:24:36.519 Philip Sellers: Yeah. And and I think it's also interesting that that you've now decoupled from your core infrastructure, which is a 1st time sort of feature. 126 00:24:37.150 --> 00:24:49.239 Philip Sellers: There's hint, hint, nudge, nudge more to come in the future. But this is this is a good use case, as you pointed out, that that serves a lot of different things. But 127 00:24:49.310 --> 00:24:55.810 Philip Sellers: we're creating tons and tons of data at the edge. Well, guess what usual suspects you can use to move 128 00:24:55.900 --> 00:25:00.799 Philip Sellers: to migrate that data from the edge into these consolidated cloud instances. 129 00:25:01.070 --> 00:25:06.609 Philip Sellers: And then you can do that in iterations and sequence those things. 130 00:25:06.910 --> 00:25:19.309 Philip Sellers: And the blog post starts to talk about all of the other things that you gain as an advantage. Now that you've migrated into nutanix files. So you've got that data portability. We talked about 131 00:25:19.560 --> 00:25:27.259 Philip Sellers: smart. VR. Also comes up as another feature. Now, what's the difference between smart Dr. And the native sort of block 132 00:25:27.280 --> 00:25:30.099 Philip Sellers: sort of Dr that spilled into these hands. 133 00:25:30.520 --> 00:25:48.790 Jirah Cox: So yeah, both can be used to solve problems right? And keep that file server available right? When things go bump or when we have some kind of a site denial style of event. We've always been able to replicate file servers themselves under the covers right at the storage layer. Send them to other clusters for recovery. Smart, Dr. Of course. 134 00:25:49.083 --> 00:26:02.589 Jirah Cox: Is almost a higher level of abstraction, higher level functionality. Where now we have running file servers on both sides that just call each other. Send their own data from one to another. Right? Versus, like, it's almost like in app replication versus like storage replication. 135 00:26:02.590 --> 00:26:21.109 Jirah Cox: So yeah, for sure, it's both are available. Both can be used to definitely solve problems and meet meet customer requirements. But that ability to have both running for even lower Rpos. Better, sla's faster failovers is huge. One use case. I'm sure we've all I know. We all met customers with this kind of challenge. Where, like. 136 00:26:21.110 --> 00:26:35.650 Jirah Cox: I have workloads at the edge. Those those sites have, you know, say, call it 10 megabit connection back to the data center. But they have a hundred megabit out to the Internet right or even faster, right? Because with a cable modem they can go really fast these days. But that might not get very fast back to my corporate data center. 137 00:26:35.790 --> 00:26:55.550 Jirah Cox: As we think about the way next generation applications are shaped right where they're going to run in containers. They're going to want data storage. How would I use Cloud effectively as Dr. For those things? Well, now, I can imagine a very consistent operation where I have containers that run at the edge. Of course I can run containers in cloud. I can have files at the edge and data services at the edge 138 00:26:55.550 --> 00:27:16.829 Jirah Cox: that can protect itself to cloud, and none of that has to even traverse my corporate data centers. Right? I can actually use cloud effectively now as edge protection with granular recovery for just the workloads it needs that, I'm not, you know, hitting that that mosquito with a bazooka just to check the box and say, yes, it's protected, right? So it's very, very right size, very economical. 139 00:27:19.000 --> 00:27:20.656 Philip Sellers: I like that one. 140 00:27:21.180 --> 00:27:25.579 Philip Sellers: I don't know that I've ever tried to shoot a mosquito with a bazooka, but. 141 00:27:25.580 --> 00:27:25.900 Jirah Cox: I mean. 142 00:27:25.900 --> 00:27:26.759 Philip Sellers: That does sound fun. 143 00:27:26.760 --> 00:27:29.850 Jirah Cox: Not recommended for for close proximity. Mosquitoes. 144 00:27:33.150 --> 00:27:52.170 Philip Sellers: you know, the other thing about that edge data, too, is is it's generally got some sort of information that we're trying to make use of. And so that's another great use case for Ec, 2, Gpu related instances where we can start to do data processing AI machine learning on that table. So 145 00:27:52.170 --> 00:28:06.260 Philip Sellers: now that it's close to or inside of an instant to cluster, it definitely makes it easier to do that. What? What additional advantages can you think of around that sort of AI Ml, use case? There. 146 00:28:06.820 --> 00:28:14.530 Ben Rogers: Well, I mean, I'm I'm personally working with a customer that's introducing AI and man this putting stress on their edge 147 00:28:14.760 --> 00:28:23.159 Ben Rogers: infrastructure. And so in some places they've looked at upgrading, the infrastructure and other places they've looked at doing like a 3rd party. What? Like a performance hub 148 00:28:23.200 --> 00:28:31.950 Ben Rogers: where they have a super fast cluster in cloud, and they're sending a lot of that out to cloud to be processed so that it's not drowning their edge deployment. 149 00:28:31.960 --> 00:28:36.809 Ben Rogers: So that's been one interesting way to see. How could we leverage the Nutanix 150 00:28:37.030 --> 00:29:02.440 Ben Rogers: product set in cloud to help, you know, build performance behind it. And, man, they're having some good success with it, of offloading those those workloads we've had. Other customers have said, man, the workloads are going to continue to get intense on edge. We need edge to do more than just vms. So now with our clusters, we're able to do. You know, AI files, traditional vms, and then we're getting into the container space pretty heavy. 151 00:29:02.440 --> 00:29:15.580 Ben Rogers: So we're able to go to those clusters that we're able to go to those customers. You don't necessarily have to increase your footprint, your hardware footprint. Let's just increase the services that are going along in those clusters. And if we have to add footprint. 152 00:29:15.580 --> 00:29:44.819 Ben Rogers: it's a commodity item, you're not having to go out and reinvent the wheel. Let's add another node, you know, less expensive nodes. And so really, 2 ways to slice. But again, the granularity of our product and the flexibility and simplicity of our product are giving our customers different ways to solve problems as they're presented as AI is coming onto the scene. We're still in the infancy of AI's deployment. So as we see more and more of this trust that Nutanix is going to be able to adjust 153 00:29:44.830 --> 00:29:47.319 Ben Rogers: and go forward with those loads. 154 00:29:47.860 --> 00:29:48.560 Jirah Cox: Yeah, so, like. 155 00:29:48.560 --> 00:29:48.910 Philip Sellers: For us. 156 00:29:48.910 --> 00:30:15.509 Jirah Cox: Customers. AI. Right now, right usually involves. Where can I get Gpus? Bring data close to that, and then execute right? That can be the edge. Because most of the data to your point. Phil comes from the edge. Right? If I'm looking at packaging inspection, you know, surveillance footage inference. The data comes from the edge. It's a matter of can I do? AI at the edge, which is always possible? Right? It gets into hardware lead times, power cooling. Do I want to run something that can run more like a data center grade node 157 00:30:15.530 --> 00:30:37.379 Jirah Cox: at the edge if you can. Great! That's fantastic. If I can't. I can think of one other way to use this, which would be, you know. How do I get data from the edge to the cloud where I know I can, of course, run AI models and make that useful right? So that ability to use nutanix as a data pipeline for that kind of use case gets very interesting, very, very fast. Whether I want to do lightweight inference on my data, or even use my data 158 00:30:37.480 --> 00:30:44.880 Jirah Cox: as as a new training data set, which, of course, will be like for training. That's a pretty much. Only in the cloud type of workload, right? Because I just need to bring 159 00:30:45.297 --> 00:30:53.959 Jirah Cox: you know. Dizzying amounts of Gpu together all at once, for a you know, a period of time, and then give it all back right very much a bursty rent versus buy type of workload. 160 00:30:54.980 --> 00:31:01.419 Philip Sellers: Well, and and then you also brought up the containerized workloads. And so that's the other benefit. Here you've got integrated 161 00:31:01.920 --> 00:31:23.540 Philip Sellers: container storage interfaces, Csi drivers for Kubernetes. Excuse me to make use of volumes and files inside of Nutanix. You've also got the object stores that we talked about, which is another cloud. Native sort of technology. So many applications are written to consume. S. 3. Compatible object storage. So 162 00:31:23.570 --> 00:31:36.590 Philip Sellers: all of those things become an advantage for you, because to to the earlier point, one of you guys made, you know, we're running cloud infrastructure at the end of the day. And so 163 00:31:36.600 --> 00:31:48.319 Philip Sellers: cloud infrastructure when we think of that as a component of lots of different services. And so you don't have to consume them all. But you do have availability to all these different services to be able to turn on and turn off 164 00:31:48.460 --> 00:31:50.400 Philip Sellers: as your business needs. 165 00:31:52.070 --> 00:32:02.259 Philip Sellers: So as we kind of trek on towards this, we'll talk a little bit more about the use cases. I mentioned object storage. There's a lot here in terms of object store. 166 00:32:03.460 --> 00:32:04.630 Philip Sellers: gives some 167 00:32:04.640 --> 00:32:15.109 Philip Sellers: information, some background again, that you know, nutanix object storage is s. 3, compatible meaning, it talks the same as aws s. 3 168 00:32:15.555 --> 00:32:32.669 Philip Sellers: but when we talk about use cases for object store. You know, we've hinted at Dr. For an Nc. 2 use case. But backup really comes into play with with object store. It's 1 of the 1st use cases, I think, for many customers. It's an easy, attainable sort of workload. 169 00:32:33.218 --> 00:32:36.330 Philip Sellers: What? What are the advantages there for running 170 00:32:36.530 --> 00:32:39.189 Philip Sellers: object store for your backups? 171 00:32:40.350 --> 00:32:52.669 Jirah Cox: Well, I would even expand it to say like using Nutanix for backups as well. Right? Because then I can deploy nutanix nutanix, unified storage right, whether that includes files for Smb. Nfs or the S. 3 for the object side of the house. Right 172 00:32:52.680 --> 00:33:13.009 Jirah Cox: now I really have an excuse me. Omnivorous ability to ingest data, right? Whatever that looks like? Is it? Is it backups giant monolithic files generated by a backup platform? Is it log data ingestion or other files that like sort of need corporate governance around them. I can just simply say yes to. However, my customers. My tenants, want to consume that. 173 00:33:13.060 --> 00:33:19.650 Jirah Cox: replicate all of that really up to Cloud. For, like that sort of honoring that 3, 2, 1 backup rule. 174 00:33:20.890 --> 00:33:31.110 Philip Sellers: I I'm cracking up over here, you know we've we've got an awesomeness button from Ben. We've now got an omnivorous button. I think I got that right, anyway. 175 00:33:31.110 --> 00:33:31.510 Jirah Cox: Yeah. 176 00:33:33.200 --> 00:33:47.019 Philip Sellers: But it is to to Ben's point earlier. I think you called it a Swiss army knife. Right? It's it really fits so many different use cases. And again, it's the same platform. And that's the power of software. 177 00:33:47.550 --> 00:33:53.609 Jirah Cox: Yeah, so that you know, I can say yes to whatever my application requires. Like I can, I can ingest the data. 178 00:33:53.640 --> 00:34:12.899 Jirah Cox: replicate it for a second copy of it, tier it to cloud. So I only hold, only hold the latest, most interesting data and other things I need to retain for a longer period of time. I can just send that somewhere else, cheaper and deeper. Let it age out there and then retire. It really becomes more like a data pipeline that I can now deploy wherever I need to easily. 179 00:34:14.340 --> 00:34:23.590 Philip Sellers: Well, and I don't think it's an either or kind of conversation. It's not a you have to use nutanix object, or you have to use Wasabi, or you have to use. 180 00:34:23.650 --> 00:34:25.859 Philip Sellers: S. 3. It is 181 00:34:25.949 --> 00:34:27.899 Philip Sellers: compatible to all. So to your 182 00:34:28.750 --> 00:34:31.819 Philip Sellers: that tearing is is an incredible, you know. 183 00:34:32.270 --> 00:34:34.329 Philip Sellers: I doubt many customers are gonna 184 00:34:34.449 --> 00:34:40.049 Philip Sellers: take the steps to do a object-based tape repository like glacier 185 00:34:40.770 --> 00:34:48.879 Philip Sellers: but that's a great place for long term. Retention to live is object fronted tape. And so. 186 00:34:48.969 --> 00:34:56.319 Philip Sellers: you know, being able to get it into this format, where it then becomes portable again. That supports that open sort of data 187 00:34:56.389 --> 00:34:59.820 Philip Sellers: migration use case for for customers. 188 00:34:59.820 --> 00:35:12.170 Jirah Cox: Yeah, you're you're cutting edge enterprise architects right now are thinking in terms of data services, not data storage systems. Right? Like, I need to offer this kind of service for this kind of application, this look in this location. 189 00:35:12.240 --> 00:35:17.870 Jirah Cox: not. I need to go buy a box with this logo on it and configure these ports and turn it up and allow these firewall rules. 190 00:35:18.000 --> 00:35:23.860 Jirah Cox: It's much more diverse and kind of fluid than that, right? So that ability to kind of follow data where it needs to go. 191 00:35:23.880 --> 00:35:32.340 Jirah Cox: host it where you need to be hosted and and maintain it for a full lifecycle. Yeah, it's fully, much more extensible than systems in the past. 192 00:35:34.980 --> 00:35:42.820 Philip Sellers: When I I have to talk about backups and immutability. Right? That's 1 of the other huge advantages to using object store. 193 00:35:42.860 --> 00:35:51.270 Philip Sellers: We're all seeking some level of ransomware protection. This is a great strategy for you. Again using common tools 194 00:35:51.660 --> 00:36:06.470 Philip Sellers: to make things happen. And now you've mitigated some of your ransomware risk. Didn't say you solved anything you've mitigated some of your risk, but you know immutable backups is really important, because. 195 00:36:06.480 --> 00:36:15.689 Philip Sellers: you know, if if you don't have good backups and something happens, you then have to pay the ransom, and none of us want to be caught in that situation. 196 00:36:15.830 --> 00:36:17.709 Jirah Cox: Yeah, at me, at the risk of 197 00:36:17.740 --> 00:36:20.010 Jirah Cox: of making a very deep statement. 198 00:36:20.090 --> 00:36:40.310 Jirah Cox: You know, the platform needs to make it easy for good people to do good things hard for bad people to do bad things right, which is kind of a duh statement. But that's where my mind goes. When when you mentioned, like the importance of like worm policy, governance, ransomware protection for a backup system. It's like I want to make it as easy as possible to write my backups. That's like, literally the job. 199 00:36:40.360 --> 00:36:50.839 Jirah Cox: But I also want to make it very hard for a bad guy to delete my backups. That's bad, right. And this is like, I know, like again I disclaimed it. This is very deep stuff, you know. It's bad to have bad guys delete your backups. 200 00:36:50.890 --> 00:36:51.370 Philip Sellers: But. 201 00:36:51.370 --> 00:37:09.410 Jirah Cox: You know. How do we make sure that that it's bad? It's hard for bad guys to do bad things. Well, these kind of policies help us out there right where, if I can have a policy that says this backup has to be retained for 30 days and can't be deleted, no matter, even if I approve it right, you need to have me and somebody else approve it before we can do that or or no one's allowed to approve that right. Don't allow it. 202 00:37:09.440 --> 00:37:32.170 Jirah Cox: That matters. That matters an awful lot. Or or you know, when I have this this kind of event happen, I want to detect it as fast as possible. Snapshot it, you know, so that I have the forensic analysis available to me. But I also want to get the business up and running as fast as possible from a prior known good, healthy state of my data. And all these things are not only possible on Nutanix. Most of them are actually automated out of the box. 203 00:37:32.550 --> 00:37:50.800 Ben Rogers: And and that's your point. There, Jaira is really key to our platform. Could you go do this in other platforms, of course, but you might be having to look at your storage platform. You might have to look at your monitoring compliance platform, and we've really built that from nuts to bolts where you can sit inside of our management screen 204 00:37:50.900 --> 00:38:06.829 Ben Rogers: and see these anomalous activities and and rectify it if you need to. And so I think the ability to monitor, think the ability to look at strange activity when it's going on, and the ability just to report back to the organization. Hey, this is what's happening in your environment. 205 00:38:06.850 --> 00:38:24.630 Ben Rogers: I'm working with a customer. That observability is a big thing for them, and age of data is a big thing for them. And we're able to basically inventory their environment. Tell them what the age of their data is, what they're doing with that data. And is there some probability that they're not using data that could be 206 00:38:24.850 --> 00:38:40.200 Ben Rogers: tiered off, or, better yet, just destroyed and gone, because it's no longer used to the company. So that flexibility there, and again our commitment to deliver those tools to our customers. I think what's really making us shine in the marketplace right now, specifically, with this, you know. 207 00:38:42.780 --> 00:39:05.230 Philip Sellers: Yeah, I agree, there's a lot of goodness. Honestly, when we started down this blog post, this is packed full. I mean, this is. Every paragraph has some new features, some some way, that you can leverage the platform for good support for good inside of your businesses and help mitigate some of the bad 208 00:39:05.567 --> 00:39:32.860 Philip Sellers: you know. There, there's so much here and again, underscoring kind of the premise of this. It's the same, no matter where you choose to run it. If you run it in Hyperscale, if you run it on Prem, if you run it with a service provider, it is the same. So I will say things are getting busy around me here at the event I'm I'm attending so I think we may end up having to wrap it up on that note. Guys. 209 00:39:33.070 --> 00:39:33.590 Philip Sellers: yeah. 210 00:39:33.760 --> 00:39:38.779 Jirah Cox: Dwayne. If you're listening, you heard he'll he'll say it shorter. Blog post, please. 211 00:39:40.100 --> 00:39:41.120 Ben Rogers: Oh, yeah. 212 00:39:41.640 --> 00:39:46.433 Philip Sellers: Well, this was this had a lot of meat on the bone, right? These were good ribs. So 213 00:39:46.830 --> 00:40:01.920 Philip Sellers: There's a lot to dig into, and we do appreciate it again. To the entire tech marketing team. That gives us good content to talk about. We appreciate you and to everybody, listening hopefully. You took away a couple of ideas of how 214 00:40:02.299 --> 00:40:13.509 Philip Sellers: you can leverage this with your hybrid multi cloud strategy. And as always, I wanted to say, thanks to Jaira and Ben, appreciate your time and appreciate you, joining us on the podcast. 215 00:40:13.670 --> 00:40:19.589 Philip Sellers: And for everybody. Listening. We hope you'll join us on another episode very, very soon.