Welcome to the Episode 338, part of the continuing series called “Behind the Scenes of the NetApp Tech ONTAP Podcast.”
We also now have a presence on the NetApp Communities page. You can subscribe there to get emails when we have new episodes.
Finding the Podcast
You can find this week’s episode here:
You can also find the Tech ONTAP Podcast on:
I also recently got asked how to leverage RSS for the podcast. You can do that here:
The following transcript was generated using Otter.ai transcription service. As it is AI generated, YMMV.
Unknown Speaker 0:00
I’m here in the basement of my house and with me today I have Anand Louis, what do you do here? I get up and how do we reach you?
Unknown Speaker 0:06
My name is Anand Louis. I’m the product manager for FlexPod products. Yes, you can reach me at an email@example.com.
Unknown Speaker 0:16
Alright, on. So you know, we’re here to talk about FlexPod XCS. But first, let’s talk about FlexPod, can you give me an overview of what that is here at NetApp
Unknown Speaker 0:24
joint solution between NetApp and Cisco NetApp, providing the oldest storage competence with the old fashioned free and fast systems. It’s an integrated solution, where we test out the different workload. So part of the flex part we have the UCS servers, the Nexus switches, and then app storage. And we also have virtualization on top so that you can you can virtualize the environment and so on. And we have part of the FlexPod we do a validation called Cisco validated design. NetApp also publishes some recommended deployments and so on to for partners and customers to leverage that as a blueprint. It’s called CVD Cisco validated design. We also test out multiple applications on top of that to SQL, Oracle, VDI, and epic, which is the healthcare applications on top of the flex part. So it gives a blueprint for partners to put it together. And they are the ones who are transacting and selling it to customers. It gives them really a good idea on how to put it together how to size it correctly, more storage, more compute. So an example would be for example, if you are deploying VDI, which is a desktop as a service offering, you require a lot of compute, and storage is leveraged just to store your environment and so on. But it’s highly compute intensive. But in other cases where customers are deploying backup and recovery as a service, that requires a lot of storage, and they leverage all the NetApp competence like CVO ONTAP, to provide the DR as a service. So it’s sold to large enterprise customers, but also mid tier and commercial customers.
Unknown Speaker 2:27
So we have other solutions, where it’s basically like a validated architecture where Cisco is not involved. So FlexPod is strictly net up plus Cisco plus validated architecture. Right?
Unknown Speaker 2:37
Correct. Exactly. Okay, exactly.
Unknown Speaker 2:41
So I understand that we also do things with one of Cisco’s products called Cisco Intersight. So what is that? Exactly? And where does it fit in with our netup products?
Unknown Speaker 2:52
Absolutely. Cisco Intersight is a SAS base offering, which started out using the Cisco UCS servers. And that became that was where they started out. And we started adding a lot more components to it. So we wanted to bring in FlexPod, under management, for customers and partners, to do several things to be able to pull the inventory to do the validation, and so on. But let’s go back to what Cisco Intersight is, it’s basically centralized management platform to first manage the UCS servers. And now we’re added Nexus, but also to deploy applications to be able to monitor your environment, to be able to give to do like cloud native services and so on. It’s a much broader platform for managing your enterprise environment, from an infrastructure point of view, but also from application deployments like cloud native. And in basically give you also monitoring in it through through I will, which is Intersight workload optimizer, which connects to all the components, looks at how it’s performing, and how to optimize your environment. So that’s Cisco Intersight. But what we did now is we actually got integrated into Cisco Intersight. First, from visibility on all storage. So we use AI QM for that to basically connect to all the clusters NetApp storage clusters, and pull the inventory up. We also added all the monitoring capabilities from on the storage site. So any alert that happens to any issues that you see, we are able to basically send it into the site.
Unknown Speaker 4:44
So there’s Intersight cover multiple different platforms into a single application, or is Is that how it works?
Unknown Speaker 4:51
Correct. So imagine you’re a customer, you have geographically distributed FlexPod, which includes that apps Storage Nexus. So these environments are geographically distributed in multiple data centers could be even in branch offices, the difficulty that customers have is to pull it all together and see it in one place. And that’s what we were primarily focused on is to be able to pull it and manage your infrastructure lifecycle management. So one of the most critical aspect of managing environment is to have an accurate inventory. So that’s the first thing we did. So when you’re deploying this environment, you need to be able to pull inventory for what storage got deployed, what compute compute, which is Cisco UCS and Nexus switches that are deployed and what how many VMs Do you have, how many virtualized servers you have? That’s the biggest difficulty that customers have. So that’s the first step is to pull inventory, being able to update the CMDB CMDB stands for config management database. And then the second is to do some level of validation, to be able to see if this environment that you have is compliant to a FlexPod standard. For that we use NetApp IMT, which is which is a solution that looks at all the hardware versions, the software versions, and say if you’re complying to some level of testing that we did. So for example, if hardware, there is a hardware and software mismatch between storage, and let’s say, Cisco UCS that hasn’t been tested, we can quickly tell you that this is out of compliance, for example.
Unknown Speaker 6:46
So I imagine it’s not just NetApp products and Intersight can handle or I would imagine, it’s, you know, Cisco, its NetApp, or it’s any other vendor out there. So it truly is one of those things that can consolidate everything into a single application. Exactly,
Unknown Speaker 6:59
exactly. So the idea that the principle that Cisco Intersight brings to the table is that it’s a single view for your application for your infrastructure to be able to manage your lifecycle. So for example, for Cisco UCS, you can do inline troubleshooting in line tag case, opening. With the we also have plans to add NetApp ASAP view that is the support view into into enterocyte. So we have customers and partners have a single point. But that’s on the infrastructure lifecycle. The next step is how do you deploy an application, so any environment that you have hardware environment, or in the case FlexPod stack, you want to deploy applications, that’s the ultimate goal for enterprise and customers to be able to deploy applications and leverage services. So for example, you want to do Infrastructure as a Service VM as a service. In our case, in the case of NetApp, you want to do disaster recovery as a service could be on the same enterprise data centers. And we use all the SnapMirror features, for example, that contact has that we had it for many, many years now. And, you know, storage efficiencies, all the all the portfolio that we the strength that we bring to the table is leveraged within FlexPod and managed through flex products. Yes. The other big value is also NetApp is natively integrated on tap, for example, is natively available in all the cloud vendors in the case of AWS is FSI x. So when you want to do DR as a service, you can pretty much say this is my starting point. This is the on tap node, or I’m starting the DR service but I want to send it to another location or I can send it to the cloud and network management management tools hides that seamlessly for you, and it monitors on your behalf. So that’s the big value added that we are bringing to the FlexPod XCS and FlexPod offering per se
Unknown Speaker 9:21
so what specifically is different about FlexPod XCS versus other FlexPods like what is this? What is what makes it special?
Unknown Speaker 9:29
So there’s two things that we would help the audience understand the FlexPod itself is continuing with the Cisco validated design. So we have written documents, where our solution teams are writing the best best practice principles in a document with XCS integration with Cisco Intersight, we are automating that basically we are giving you management tools such as pay least. But for some customers who are in dark sides, for example, you could be that connected, it’s an appliance that can be connected or completely private appliance is basically bringing the management layer on top of that. So when you have a very complex environment, multiple geographically distributed environment, the difficulty is that IT admins have the infrastructure team is how do I manage this? How do I pull the inventory? How do I monitor it, so that I’m not spending all my cycles watching it, but it’s, it’s basically lights out monitoring on behalf of the operations team. This can also be integrated into existing monitoring tools, for example, so that the centralized NOC can receive alerts on behalf of the operators. So it’s basically doing that. But also insight into your compliance. And in the future, we want to be able to insert any what we call proactive alerting. So if there is a P cert, if there is any type of security vulnerability that we detect, and that app is excellent at that with ASAP, where we proactively tell the customers at these issues like here, we’ve seen it multiple environments, this is the recommended patch. This is the recommended fix that we have. And that app has been doing it for years now. That’s the same principle that we want to bring to FlexPod XCS, to that feature of proactive alerting is available for Cisco UCS. So TAC and pro Cisco TAC and proactively alert customers, that same principle we want to bring in the future with XCS on the net upside.
Unknown Speaker 11:50
So it’s a proactive alerting, is there any sort of functionality where I can just click a button and fix problems or click a button and upgrade things like that? Or is that something that’s just not possible today?
Unknown Speaker 12:01
On the UCS side, it’s possible so on the Cisco UCS, because it’s integrated into into into UCS with Cisco decide the troubleshooting repair path, upgrading recommendations are available. We have it independently with AI key one on our mat, Cisco on the network management tools, where we can proactively tell you what is going on and also recommend a path but that’s the recommended upgrade. We want to also add it into Intersight storage admins can do it manually with a IQ and management tools. But that’s that’s the feature we want to add into Intersight later.
Unknown Speaker 12:48
Okay, XCS is the name of it, what exactly does that stand for? Like, what is the abbreviation there?
Unknown Speaker 12:55
It’s cross cloud systems. Okay, it’s FlexPod cross cloud system.
Unknown Speaker 13:01
Okay. So when we say cross cloud, that implies that we have multi cloud hybrid cloud involved here. So it does it support all the major cloud vendors are just a few of them, is it this is support like CBO? And CVS? What all does that include? We have tested
Unknown Speaker 13:18
and validated with AWS on a sec. So basically, doing the hours of service from on prem to AWS with the the NetApp hybrid cloud capabilities are what we are bringing into FlexPod. So all the native integration that we have, so native integration means this right the NetApp storage efficiency that you have on prem Data ONTAP software, can is running on Azure is running on Google is running on AWS. So that efficiency that when you take a snapshot, you’re using a lot less storage, from native AWS storage, for example. That’s what we’re bringing, to to answer come back to your question, what are we bringing to the table is that hybrid cloud capability that NetApp has, with all the management capability, like CVO is what we are planning to add in the future. But it’s going to be validated, tested, and documented. And then automated through XCS later,
Unknown Speaker 14:26
what sort of story do you have with things like Asher or Kubernetes? Is that included in any of this? Is that something that’s down down the line there? What’s, what’s the involvement there?
Unknown Speaker 14:38
So NetApp and Cisco we have both of us on the on the Cisco side compute with AKs automation on the on the cloud native side, and then NetApp has Astra so we have developed a CVD at Cisco validated design that explains how these two technologies open source cloud native technologies start working. And we publish the CVD. But the plan is to also automate it through through XCS. So the idea is that as we are building these services into the CVD, we want to provide that as a service to the customers. So the initial where we standing today from a feature, or technology viewpoint, is infrastructure management, the visibility portion of FlexPod is, is GA, we announced it on July, July 14. But the next phase is definitely the automation the consumption layer. So the most critical part when it comes to enterprise customers, is to offer it to their end users. So you’re buying this hardware for a purpose to offer services to your internal business. So if you if there is a bunch of developers who want to develop an application, for example, on SQL or VDI, or just, they just want a VM, that service needs to get automated. So they do that through ServiceNow integration through Ansible. scripting, with the the idea with FlexPod years is to bring these out of the box automation capabilities to the customers. So I’ll give you I’ll rattle out a bunch of services that it when we talk to partners and customers are doing today. And that’s what we would like to automate in XCS. It so VM as a service infrastructure as a service. But from a storage viewpoint, they typically do DR as a service. And there are multiple grades up that right you want to do active active, you want to use VMware Tools like SRM or for example, application aware backup and recovery. Those are the services that customers are doing today. But they are putting it together using NetApps, Ansible scripting, but the plan is to put it out of the box so they can easily consume, modify the workflows and make it their own. The idea is also to push it to partners to get adopted. So we want to build a automation practice outside met up with the partners.
Unknown Speaker 17:25
So what does this automation piece is going to be driven by is going to be like an Ansible thing? Or is it going to be like a Kubernetes service? How are you going to handle that portion of the automation?
Unknown Speaker 17:36
So XCS itself has been it says go into site has what we call a orchestrator Cisco Intersight orchestrator, a cloud orchestrator, which has built in connectors from NetApp. And it has a bunch of tasks to do atomic operations on storage level. So we can put together workflows specifically on run storage automation, like bring up what we call day one automation, where you’re expanding, shrinking, storage, allocating more, creating a cluster, adding a node to the cluster. So these services are already feasible. So we have two second approach also were net up provides a full set of open source scripting capability with Ansible. Today, and there is a we’re looking at TerraForm in the future, where you can do all the NetApp services on top of let’s say cloud inside see Cloud Manager, or CVO AI QM Alder on prem as well as cloud services through Ansible scripting. What we’re looking to do with the Cisco Intersight is to provide a bunch of reference workflows. So that these are references in terms of how to do the or as a service how to do V as a service, and provide it as a reference implementation, but let the partners and customers adapted to their environment. So each customer environment is slightly different. But typically the ITSM architecture in a customer, I’m talking about enterprise guys, the they have already designed and looked at how to do centralized CMDB. How to do CMDB is basically your inventory update, very accurate inventory on what you have in terms of on prem assets, compute network storage, that modeling is built into the CMDB but also your cloud assets abroad and so and then they they lay on top of that automation workflows. So depending on what type of service you want to customers want to offer to the end users. They tie these, what we call service catalogs on top of these workflows. So the idea is to provide reference workflows in in flux products. Yes. Did that answer your question? Justin? Yeah,
Unknown Speaker 20:09
just was curious, like how we’re integrating with the automation piece. And that sounds like it’s already there. It’s just a matter of building the hooks.
Unknown Speaker 20:16
Correct? Exactly. And building the reference workforce. So the idea is to, for, for the solution teams on both sides is to provide reference workflows, and to help the partner community build the automation practice, from and since jointly built between Cisco and and that that’s, is going to be stronger in terms of adoption.
Unknown Speaker 20:42
So I imagine that something like this isn’t for, you know, home directory shares, or like, you know, basic NAS sharing, there’s other use cases here that are more, I guess, apt for this type of architecture, right? You know, you’re not going to buy an entire box set of something when you’re just doing simple things. So what sort of use cases are you seeing out there for XCS? Like, what are the main goals here for that particular product.
Unknown Speaker 21:10
So the main goal is to help customers, commercial customers, or enterprise customers, get control of their FlexPod environments. So with with the customers have been telling us and partners as well, it takes a heck of a lot of time for them to go, run the scripts, collect the inventory, and see if this environment is in compliance or not. And it takes them a lot long time, it’s a paid service. So every time they need to upgrade an environment, they run these audit services, right. So the idea is to provide it to partners as a managed services offering that they can put together. And also to the end customers, where it’s at their fingertips collection of the inventory, update to the CMDB, running the IMT compliance, and monitoring of your environment is is basically done de facto. And in the case of UCS, it’s going to be in line support. So the advantage is, so that’s the primary focus right now is to strengthen the visibility, bring in all the inline tag supports that slightly in the future. To bring control of your environment, it’s huge, actually, several partners told us that this is great going to save them huge number of hours, in terms of bringing the environment into compliance. That’s the first phase. And we want that to be successful. The visibility pillar of flextight XCS. The second aspect is we want to focus on automation. That’s what we are embarking on with Cisco in the next phase.
Unknown Speaker 22:57
And I guess maybe what I was getting at was what sort of workloads fit best for this? I mean, I don’t think you’re doing this at a small scale. It’s got to be larger scale stuff. Right? So what are people looking at this for?
Unknown Speaker 23:09
Oh, so in general, I’ll come back slightly. In general, for FlexPod. We have small commercial customers who buy it, because it’s convenient, it’s put together by partners. So depending on your workload, partners have the ability to stretch either more compute, more storage, and it’s literally catered to their environment, on the hardware design. And then if they want to stretch it and put it in, in the cloud, for commercial customers, the setup is done manually. You know, they run some scripts and NetApp and Cisco provide the zero bring up of the flex box. So that’s VM as a service. You know, you want to you want to VM and then we want to set pick the operating system and so on. But as you go into the enterprise customers, specifically, let’s say health care, or, or slightly bigger enterprise, guys, they want to look at workloads such as SQL. So for example, the most popular workload that we have is SQL, then followed by Oracle, we definitely have VDI, SAP and epic applications. So the automation fits really well when you have often demands coming from the developers. But automation is starting from an Arab point. ders services the most popular because they need to make sure that the data is protected, efficiently stored remotely for disaster recovery, for example, that’s the most popular service from from a NetApp viewpoint. But from application consumption viewpoint VM as a service is the most popular and then followed by SQL. And then for customers who have, who have SAP as their internal CRM, that’s next. And then you have SQL as a service.
Unknown Speaker 25:16
So sounds like you can make this with different sized platforms, right? Like, it’s not always gonna be an 800. It could be a fast system, right? Is that what I’m hearing?
Unknown Speaker 25:24
Correct? Exactly. So the advantage with FlexPod is that it was designed to, to scale from the smallest workload, smallest environment, let’s say 200, VMs, all the way up to you know, 20,000, VMs, and so on. It’s really built to scale, more compute, you can stretch it into different X, Y axis, more compute or storage, depending on what it guys have to build internally. And we have these hooks into all these hyper scalars. That makes it even easier. So we heard from multiple customers, that our tools makes this automation so seamless. From a NetApp viewpoint, we just hide all that complexity. So for example, they want to do DR as a service, they just picked a source a target, and then they say, here’s this map Snapmare relationship, they hand it over to CVO, it’s automatically monitoring on their behalf. And they said, it’s that easy, we need to tear it down, remove it, it’s simple, we just go send a Delete on the at a top level relationship. And the tool takes care of what we heard also from partners is that this consistent software did what we call the NetApp Data Fabric is so consistent in terms of offering for a number of years, it makes it easy for them to build that practice around around around that. And that strength carries over to FlexPod.
Unknown Speaker 27:05
So on and you know, it’s all well and good to talk about hypothetical use cases. Do we have any actual use cases out there in the real world with this? Do we have any customers that are already spun this up?
Unknown Speaker 27:15
Yeah, absolutely. We have a federal customer, public sector customer who has deployed Astra, so they basically they’re using the Astra services Kubernetes services to be able to deploy storage in on prem or in the cloud seamlessly. So this is dedicated Gulf cloud, where using the automation scripts, where we can deploy on prem or, or in the cloud seamlessly. So that’s the major advantage of what we built in terms of Kubernetes services for storage.
Unknown Speaker 27:57
Okay, so it sounds like we’ve already got some people that are using this and probably lots of others that are kicking the tires on it, and they’re interested in it. So hopefully, you know, if people are interested, they can find more information. So do you have any sort of places people can go to ask you for more information?
Unknown Speaker 28:12
Yeah, absolutely. We have published a CVD that we will put in the in the speaker notes, where you can find the reference or you can also reach out to us to get more details about what we did, how the customers did it. We can’t divulge the name, but we can tell you more details, technical details on what they did.
Unknown Speaker 28:35
Okay, so it sounds like we got a lot of good places to start here. If we want to learn more about FlexPod XCS. And like we said, we’ll include a lot of this information in the blog that goes along with it with some links. Again, on and if we want to reach you, how do we do that?
Unknown Speaker 28:48
You can reach me at firstname.lastname@example.org.
Unknown Speaker 28:56
All right, excellent. Well, thanks so much for talking to us all about FlexPod XCS and letting us know what’s out there.
Transcribed by https://otter.ai