Voices in Cloud – Episode 18: A Conversation with Jeff Dickey of NetApp

Guest

Jeff is a forward and creative thinking leader in technology and business strategy, product development and partnership relationships in the enterprise technology and cloud space. During 20+ years of work experience, Jeff has built a strong track record for identifying and nurturing emerging technologies; these companies often develop powerful market adoption and/or end in successful exits. He has a deep understanding for what it takes a company to accelerate the adoption of new technologies to transform their business, drive ROI and compete in a new marketplace.

As an exploratory technologist who is an expert in evaluating products for short- and long-term fit, he excels at technology product creation and GTM strategies for successful market adoption. Jeff looks to the bleeding edge and productizes what can translate to the enterprise. He also brings key relationships with VC’s, industry analysts, thought leaders and news organizations. Jeff has personally emerged as as a thought-leader in cloud, speaking at cloud conferences around the world, doing guest appearances on popular podcasts, and hosting the popular Nextcast podcast series.

Transcript

David Linthicum: Hey, guys. Welcome to the Gigaom Voices in Cloud podcast. This is the one place where you can hear from industry thought leaders providing no-nonsense advice on how to succeed with cloud computing, IoT edge computing, and cognitive computing. I’m Dave Linthicum, best-selling author, speaker, executive, and B-list geek, and joining me today is my special guest, Jeff Dickey. Jeff is a forward and creative thinker, leader in the technology and business strategy space including product development and partnership relationships in the enterprise technology and cloud space.

During 20-plus years of work experience, Jeff has built a strong track record for identifying and nurturing emerging technologies. These companies often develop powerful market adoption and/or end up doing successful exits. That’s always a good thing. (Yeah, and get your money out of there.) He has a deep understanding as [to] what it takes for companies to accelerate the adoption of new technology to transform their business, drive ROI, and compete in a new marketplace. He has 10 years of cloud experience, 15 years of consulting, and spent several years on technology advisory boards for Intel, Dell, VMware, and others. He’s been on a few different podcasts including GeekWire and OpenStack podcast. We won’t hold that against him. How are you doing, Jeff?

Jeff Dickey: I’m doing great. How are you?

Anything we missed going on? What are you doing these days? I think we used to work together, correct?

Yeah. I was trying to look it up. I know we were on each other’s podcasts in 2014, but there was one even before that, but I couldn’t find it.

Yeah. I think this is my fourth podcast in the last – hold on, there was SOA [service- oriented architecture] podcast, there was a cloud podcast, and SOA podcast was InfoWorld. There was a cloud podcast, also InfoWorld. There was my podcast when I owned Blue Mountain Labs, and there was a podcast when I was at CTP. Now lots of other podcasts with current employers and things like that. It’s a really good way to spread the word because you and I will spend about 20 minutes having a very easy discussion without hemming and hawing over what we said in the words. In doing blogs, I have a tendency to think that’s too slow. The market’s moving too quickly. We have to get the information out there quickly. You’re with NetApp now, and so what do you do for those guys?

I’m spreading the good news on our cloud products. I think a lot of folks don’t know that we have, not only a cloud product, but a plethora of cloud products. Most people still think of us as enterprise storage in the data center.

That’s the legacy, right? All these companies have to change their stripes and modernize for where things are going. What’s the exciting news around NetApp in the last five years?

It’s interesting. NetApp has actually created more products in the last 18 months than NetApp has in the last 26 years since it started. NetApp was a very big player in the early dot com days, right? Really powered the internet. It was Sun, Solaris Systems, and connecting to NetApp, but because we’re creating these incredible cloud products – and I think what’s great too is we’re using our own cloud products to develop and innovate faster – we keep turning out more great products to be used in conjunction with the cloud world. We still have a very, very large on-premise world, right? I think we haven’t even hit that tipping point of the enterprise critical apps.

My assertion is always that we’re going to hit a saturation point even if we hit a tipping point. Right now, we spent the last five or six years getting the low-hanging fruit into the cloud: things you can typically ‘lift and shift.’ Java-based applications, Python-based applications, LAMP Stack-based stuff, things like that. We’re getting to things that are very much harder to do. Things that have to be refactored and changed, and data sets that have to be redone, restructured as we move them into the web. I do think we’re going to hit a point where it’s just [not] economically viable to move certain data sets and certain workloads into the cloud.

I think those things will remain on-premise. It’s funny, it’s like everybody considers me the cloud guy, and one of the stories I tell is that I think that one of the better businesses to invest in right now would be what’s going to happen to those 30 - 40% of the applications as to where they’re going to run, what kind of hardware they’re going to leverage, things like that. I think the big enterprise companies, NetApp included, are trying to look at where the next generation of this stuff is going to be. Am I missing something?

What I’ve been talking about, even for the last few years, just from doing cloud for so long and helping customers, even from the beginning days of getting those low-hanging fruit applications or new development out in a public cloud, what I think has really happened is: the longer you’ve been doing this, moving to public cloud, but also maintaining your on-prem, you’re creating this large chasm between your environments. The longer you’re doing it, the greater the chasm is. You’re not only managing one in cloud, one on-prem, and then one together, right?

That’s three infrastructures, but you’re putting more money and more time into public cloud and the legacy stuff is often neglected, and often getting harder to manage because you’ve put so much work into your cloud-native apps and to the automation that everything provides. It’s getting harder and harder to move these business-critical apps. Apps that really generate the most money for a lot of these companies, especially in the big enterprise and the Fortune 500 companies. We all know[that] the more data sets you get together, the better your information is, right? The better the quality you’re getting. We want to get that information together. We want that seamless across anywhere.

And you guys have experience in doing it, so probably more than some of the other companies out there. What should some of the millennials who just got into cloud computing for the last five years and haven’t experienced the enterprise-based drudgery that we both have in our careers, what should they know about NetApp that’s a key value proposition?

I would think of NetApp today as data management and data governance. There’s a lot of enterprise features that you may not be aware of if you’ve started in cloud or born on the cloud. You’re not really familiar with snapshots. There’s some types of data replications, but you’re not really replicating encrypted data. You’re not maintaining point-in-time copies and giving maybe active directory credentials and permissions to multiple different groups and maintaining that. It’s really about how are you managing massive data sets, the security around it, and the connectivity. It’s NFS, right? We’ve got multiple protocols going on. We can handle Windows and Linux.

It’s quite different than what you’re used to in cloud as far as ‘Hey, I’ve got to figure out what a

I’m I actually putting on disc in cloud, right? How many iApps do I need? What size do I need this?’ If I’m looking at block storage, there’s a whole bunch of things you have to look at. What happens when you need to grow that? What happens when you need better performance on that? It becomes hard to manage [with] a lot of choice. In the enterprise world, you throw that all in NetApp. It’s a shared volume. You usually throw millions of dollars at it to get the performance, but you don’t really have to think about that. That manages the storage.

That’s where Azure NetApp Files fits into the mix?

Exactly, yeah. Now Azure NetApp Files is a first-party service from Azure and it’s in your console, but it’s a little bit different storage. It’s not meant to be the end-all of storage. It’s meant to complement and it’s meant to be used in addition to what you’re already using in Azure. If you’re moving SAP over, the best way to do that is on shared storage. You don’t have to really predict what that bandwidth that you need or – how many iApps. You don’t have to think about these plumbing, storage infrastructure details. You move all your data or you copy all your data with some of our other tools and you’re off to the races.

Now later on, when you say, “Oh, I need more performance off of that. I need some serious performance on my data.” It’s an API call, or you’re the UI and you’ve changed that to a high-performance tier. If anyone’s ever done that in public cloud, creating new storage and copying and getting everything connected back over to those volumes, it’s pretty difficult, and we’ve solved that. Again, it’s complementary, right?

What other products are you guys pushing in the market that are very much aligned with the same strategy?

Another one of my favorite ones is called Cloud Insights. It’s taken from a product that is probably in almost all of the Fortune 100 companies that was a very big lift to bring in, and to look at very deep infrastructure and application performance and metrics, which we turned into a cloud product, a SaaS product, and it can look[like] basically anything that you point it at, right? It’s not just infrastructure. It’s not just application. It’s everything, right?

You can see deeper performance and what’s going on and bottlenecks of Kubernetes and/or storage and/or VMs. Great for networking too. Again, I like to tell people, we have all these cloud products and they are fairly new, but they’re production-ready; they’re being used at scale. Right now, we’re still adding features to them. They don’t go head-to-head with some of the best out there, but they will. We have a portfolio of them. It’s almost like you can do less vendor management with our portfolio.

What are the use cases out there that people are leveraging your technology for?

The ones I’m seeing the most right now being implemented and questions asked [about], [I was] just on a seven-city tour and getting asked about again, moving SAP. People are really trying to get SAP moved into the cloud. Also, it’s Oracle and some other big databases that are on-prem. People that have large Linux Farms are looking at this because it’s a single-point storage for both – enterprise companies have a ton of legacy Windows and even new Windows, but they also have a lot of Linux, and they’re trying to manage that as well. HPC is another one that we’re seeing a lot.

I think the most interesting thing that I’ve seen is the folks that are using our products for HPC or big data or even large Linux Farms, is after they see what they can do with that, with those use cases, now they’re trying to figure out what else they can do with it.

What are the two most popular use cases for your technology?

Right now, I would say HPC and SAP, which are two of the hardest ones out there. It’s really hard to satisfy anyone’s requirements and goals when you’re talking about just the performance it takes to do a large-scale SAP or even the big HPC. HPC storage has always been – you’ve always have it distributed. The more nodes you throw at it, the better you’re getting, and we solved that.

I remember watching five-year install times for SAP or three, so I think people run scared when you have to do anything with that technology. Would you agree?

Anything, anything, and not only that, what else – when you get into it, it’s not just SAP, right? It’s what else have you integrated into your SAP environment? What are the homegrown stuff you’ve put in? What are the third-party pieces that have been added? Again, most SAP installs are not brand new. They’re very old. Most people who did install and know the keys and know the skeletons everywhere are long gone.

Going forward, this is the harder problem like I mentioned that we’re looking to solve. We’ve been babied a bit with cloud-based systems that are solvable with very simple ‘lift and shift’ kind of things. It is moving the ERP systems, big data systems; the existing legacy environments have to run some place, and the ability to not only move them into the cloud, and I think in many instances that’s going to be contraindicated with the ability to modernize these things. Then we’re going to be in a much more efficient state. In fact, I think we should focus on modernization of the stuff versus relocating everything into the cloud. What’s your take on that?

This week in Dallas, I had a very long conversation with someone who runs a 20,000 ESX host. We’re talking about migrating this. I feel I’m a little bit different. I think lift and shift is not the best way to do things, but I’m very much a fan of lift and shift because when you go in and try to refactor on-prem, you still have a lot of complexities, and you’re still chained with your infrastructure.

If you can lift and shift – one, I’ve never seen anyone need more resources than they had when moving, right? You’re always going to smaller instances in cloud, and you’re now – again, the chains are off. Now you can start refactoring and re-architecting inside the cloud with everything at your disposal. Even with enterprise apps, you may not be able to scale out. That might be hard, but there’s a lot of ways to scale up, and there’s a lot of ways to automate. You can still automate legacy applications when you’re on the cloud.

What are you guys doing in the world of Kubernetes?

Kubernetes, that’s a very, very interesting piece there. It’s why I’m over at NetApp as well. NetApp has been buying some of my favorite cloud companies. One of them was a company called Stackpoint, a company – if you wanted managed Kubernetes installed anywhere you could think of, they have connections everywhere, on-prem and in the cloud, that’s who you would use. NetApp bought Stackpoint and now has everything integrated into our other cloud products and to our on-prem products. It’s called NKS, NetApp Kubernetes Service. Kubernetes is hard. It’s very much infrastructure. Its infrastructure as code, but it’s still infrastructure.

When I talk to people out there and I spend a lot of time and meetings at Microsoft Build, one, most people are like, ‘what’s Kubernetes, what’s NetApp? Two, anyone that was using Kubernetes that was at the conference was using a managed Kubernetes service with AKS. It’s very difficult and it’s very difficult to scale it and to do upgrades, but your developers want that. They love the flexibility. They love what they can do with it. We’re seeing that we have a lot of customers that want apps in several different locations. They want their Kubernetes on-prem and in a couple of different clouds. The apps may not be able to federate quite yet. We’re still working on developing federated apps, but we’re there. We’re there to support it. You can build federated apps now. The plumbing is laid down.

What advice would you give to an enterprise that’s looking to move substantial application infrastructures into Kubernetes?

One, don’t do it alone. If you’re going to ride a bike, you don’t get on a motorcycle and take off, right? You get some training wheels on and you do that. I think that’s what we offer. It’s again, fully managed. Everything’s connected. It’s a very simple UI and a great API access. We have trials that you can put your apps on it and try it out and let it run. It’s also one of those things where yes, the training wheels can come off and it is a sport bike if you want it to [be]. It grows with you. It’s probably the best starting point that you can have, though.

Yeah, I think it is a pretty good starting point. What do you think the evolution is going to be over the next five years as far as use for Kubernetes? What are we going to be talking about if we’re doing this podcast in 2024?

That’s interesting. Especially for us talking, right? I think on the first podcast I was on with you, we were talking about CloudStack. I know I had you on my OpenStack podcast. Now we’re talking Kubernetes. I think Kubernetes is going to live longer though. I think Kubernetes has a very long life because of how it’s been set up, the adoption of it. There’s so much room for growth and it’s done the right way, and it’s integrated into what everyone’s doing as far as – every vendor out there is integrating it the right way and is using it. It is the best way to do cloud native applications.

Yeah. It’s not trying to replace the cloud. I think that’s where we run into it and unless you have $20 billion to spend on R&D and marketing, that’s going to be a very tough market to break into. What we’re doing is, in essence, trying to augment the platforms within platforms, correct?

You’re right. You’re right.

If you look at Kubernetes, it’s bringing to bear a lot of the things around the to-do list for lots of folks who were doing the next generation architectures. Certainly, when we developed containers and you say, “All this stuff is cool. We want to build applications this way, but no way we can make these things scale and no way can we be leveraging using clustering and orchestration, scheduling as well outside of something like Kubernetes.” That’s why I think it just exploded in use. What kind of new innovations do you guys have going on at NetApp going forward?

Oh, there’s so many things. I think the best thing that excites me the most is a product we have called Orchestrator. The best way to explain this is: imagine using tags for your data and your applications. If you created a tag called ‘production’ and assigned it all the values that you thought were necessary, like you wanted to run it in these many places. You wanted to have it run these many containers or this type of persistence that you would want. You want this type of replication or this many backups and where it goes. It’s thinking about the applications and data together. Anyone in your infrastructure, if you’re using Orchestrator, and let’s say you were launching Kubernetes, and you would tag those applications as production and maybe tag a couple of data sets as production, that’s it, right? It’s following these rules.

We’re really simplifying what you’re doing with – again, this very complex world of multiple clouds, and on-prem and hundreds of vendors, we’re making this very simple so that everything is visible and everything is tagged and everything is automated because of the structure you put in place. You can search the tags. You can look up everything. Imagine your apps deploying from where your data is and your data is where your apps want to be, right? That’s the future we’re building right now.

Where can we find more information on the web about NetApp and yourself?

Go to cloud.netapp.com. That’s a great overview of our products. It’s also the portal that you would log into and to demo everything that we have. You can find me, I’m usually on the Twitter @jeffdickey.

Go check out Jeff’s Twitter account. Like I said, he’s in networking and a mover and shaker in the industry as well. Also, look at NetApp in terms of their various technologies that they have to offer. Folks, who are just getting into the cloud space that don’t understand where NetApp came from, [it’s a] very strong player in the enterprise space and they’re taking that to the cloud as well.

Please pick up a copy of my book, Cloud Computing and SOA Convergence available on Amazon and all places books are sold. Also make sure to follow me on Twitter at @DavidLinthicum, as well as LinkedIn where I have several cloud computing courses on LinkedIn Learning.

Until next time, best of luck in building your cloud computing offerings. Talk to you guys real soon. Bye-bye.