NetApp HCI (Hyper Converged Infrastructure) Tech Talk

[Music] hey everybody it's Chris Powell with champion Solutions Group today I'm with West Talbert principal architect cloud infrastructure with net a thanks a lot for coming back to you thank you really excited about the new announcement about the end you know the NetApp HCI really exciting about the you know the announcement the architecture everything I've been reading about it one of the key things I think that we're really interesting for you guys first foray into this market you know Gartner named you guys to nap HCI is an evolutionary disrupter not bad way to come out right out of the chute so can you maybe tell us you know what about this evolutionary disruptor what makes us a lot of evolutionary sure so what makes us so loop evolutionary's if you look at all the hyper converse platforms they've all gone with a shared core architecture between the storage of the compute that challenge around that is at enterprise scale you're going to have performance issues so within a different path what we did is we actually separate it we're computing storage use separate cores within the architecture this allows us to protect the performance of the environment overall and have better growth not to mention lower licensing cost over time scale mm-hmm that's important especially the lower licensing cost it is I got it um so just you know upfront what's the key difference though would you say between what you guys just came out with with HDI you know versus the others in this space right now what do you think the key differences are well you know one of them would be our quality of service so NetApp acquired SolidFire in December of 2015 you remember that assertion about 870 million dollars and what SolidFire zelma no less gives us is quality of service via 10 Giga I scuzzy at the individual volume liar so for us what that means in terms of performance for NAT fhc is our ability to guarantee performance for all the workloads right and so there's actually a white paper we put out I guess it was October or November of 2017 where we put a BB I MongoDB and Microsoft SQL so two different types of databases plus BT I on the same conversion the structure on that ACI and we were able to keep the performance appropriate for everybody within all those varying workloads and that makes it in in the tenon environment all together perfectly running hmm now here's the catch okay most of those other HDI solutions in the industry heck all of the other HDI solutions in the industry they don't do that at all what they'll do instead is little employ separate islands because they can't holistically let these mixed workloads reside together do the performance issues that will occur so it's really exciting for us to be able to give the customer true enterprise scale HCI that supports that multi tenant mixed workload environment hmm absolutely excellent another question that keeps coming up you know that we see in the marketplace so with your other competitors will not gonna name them but you know some of the challenges that we've heard is that hey I'm running out of compute when they have to buy a whole new stack that's right how is yours different which I think a lot of our audience is really gonna appreciate this how is your aunt where instead of having to buy a whole new stack you know you just need more compute you know how's yours different so there's a few key points there right if you look at what Gartner seven they call this an evolutionary disruptor a lot of that had to do with the fact that we have the stores in the compute course physically separate and what that means for us is that our customers can purchase storage or compute and then scale it independent within our architecture now what's further if you look at the compute node offerings within our solution and the stores node offerings within our solution both of them have a mix in terms of I ops to capacity own storage side and memory two cores on the compute side so it almost feels like for the customer as they add the compute in the storage nodes they're scaling these four variables completely independent right because of those mixtures therein so it's really exciting from that standpoint it's a part of how we achieve achieve true enterprise scale but what's more and this is very interesting what if your customer today they're in the situation where they just refresh their computer sure they just refresh their compute hopefully they just chicken pupae champion all right the tan those those compute nodes in there in there compute environment for the virtualization let's say they're 18 to 24 more months before they're fully depreciated was a result of that they can't toss them out early into a Greenville hci'm deployment but with that HDI you can implement an nfh see I just do to compute nodes within the initial configuration do the rest is all storage present I scuzzy 10 gig volumes guaranteeing quality of service on every one of those volumes to the external hosts that they purchased from you previously they're not fully depreciated right so what they end up with is your legacy compute we'll call it a legacy compute I the two new compute nodes within the new infrastructure the shared storage therein and then as those nodes from their previous implementation begin to fully depreciate from an accounting perspective right so this is a business conversation you're not a technical conversation then you can add more compute and you simply vmotion the compute side over obviously the stores is already there so what it means is if they've just bought all that compute from you let's use an NF HCI as the storage in drove implemented more storage focused than compute focus and then over time we can slowly mind migrated over which is a beautiful conversation I have it's much different than these greenfield conversations that all of our competitors have in the HCI space which is like sort of like a rip and replace type of motion that's right and nobody wants that right if I buy the compute I want to leverage the compute as far as I can that's really really interesting so you know speaking from the business perspective it's great technology we talked about that but you know for the audience what workloads do you see HCI really dominating it way what business cases do you see HCI really you know doing well in so for us specifically is net HCI our ability to do multi tenant mixed workloads and guarantee the performance for everyone means that we can be that one-stop-shop for that private cloud type of market where you bring in all these tenants and that's great for us much different than the other solutions where they'll implement an island strictly for VDI then a separate island for general compute then yet another Island right that they may use for their tests and yeah so we don't want to put you into a point solution you can actually guarantee the performance no matter what the workload and around multi-tenancy that's right at the same time so it's very unlike others who talk a lot about well yeah this is a great point solution by a spec for this vbi solution or yes for this you're saying with your technology that you know basically any business case are you gonna run like a high OLTP you know online transaction processing the database off of it as well so you know potentially you could but and this is a part of what's so exciting about being a part in that specifically you look at some of these other players and they're a one-trick pony this is all they've got to pitch right right but for us it's different it's beautiful because what if somebody comes to me and they say hey I need a LTP but I need a million I ops on a single volume well in that situation none of the HDI solutions none of ours you have to be the right one right but for us at naina because we have this holistic this entire portfolio that we can leverage we can bring something else to bear for them whatever they're looking for right yeah I gotcha yeah so let me just ask you I know we spoke about this we're just you know coming to the end of this interview here but I was reading how easy this thing is to deploy what no I mean how easy really is it to deploy it is easy so for us the deployment process usually looks like about 45 to 50 minutes typically what you do is ahead of time we'll send the customer this spreadsheet where you go through these different IPS and subnet information you'll collect all the details ahead of time after you get everything rack stack and pin you go through a menu and you punch in all of those different IP addresses all the DNS that sort of thing do I need to build a fresh V Center am I going to join existing vCenter and then you go through all that you say looks good let's go come back 45 minutes later you have a fully functioning turnkey nav HDI solution running VMware vCenter with the motion enabled right out of the gates for lazy so it comes with a nice cookbook there if champagnes here to help of course to you know get the things set up and implemented properly now it's installed up runnin what does it take from a management standpoint how streamlined is it have you brought everything together from a single neo management perspective how is that so for us the management is great because what we've used for the management of NetApp HCI is we're actually just using the SolidFire be Center plugin and this is important the SolidFire V Center plugin that we've had around for many years to manage SolidFire in any existing VMware environment the reason I bring that up is of course we're in the first nine months right since our shift day but the plug-in that really orchestrates everything has been around for years so we know that we can trust it we know it functions a great at scale and what's more is our ability to plug in with other solutions so case in point we can backup to s3 buckets and the cloud natively with that epics yeah nice we can deploy on tap select as a VM into a net up a CI picture if you will you've got a call center you need a small data center box that call center so we implement a small net up API for that but within it we have it deploy a venous on tap select as a venous within it so we've got local Nazz capabilities there at that call center but then you point that on tap select it's running and then that F HCI back to your on tap investments your primary data center which means that this is not some Island of a solution it functions with the rest of the customers investment as they bought fast and AFF any on tap based technologies three champion and deployed an elsewhere right so it's not just a one-trick pony if you will and if this was a customer's first foray into the net apps family from a management standpoint they didn't have any of the on tap stop in the past maybe this is their first foray into it does the management management tools work will work with any of their existing management tools that a standalone does that work so all of our management is within these centers okay so we didn't elect to push for a separate manner smart so yeah keeping it all in the center that you know and here's the real beauty you know I managed people for a point in my career in past what I really found was hard is when you've got a backfill an opportunity backfill in your organization and you're looking for a specific skills that's really tough well for us because all of our management is native within B Center and at this point in the in the if there's any market is everywhere right it's easy for me to find somebody to manage that environment as opposed to some niche smart console that they've got to be trained on yep right not to mention that increases their flight risk to go somewhere else after you invest them having trained right trussville we've heard that I do feel you on that well listen it's been great looking forward to you know getting more engaged with it and everybody out there thank you very much for tuning in and look forward to picking you next time thank you so much brother thank you

Channel: Champion Solutions Group

Views: 1,102

Rating: 4.963964 out of 5

Keywords:

Id: bo-Trhydyz4

Length: 11min 43sec (703 seconds)

Published: Thu Apr 26 2018