Advancing Life & Work
HPE-Editor

HPE Tech Talk Podcast - New GreenLake Lighthouse: A new Cloud Platform that defies convention, Ep.18

Many Enterprises believe there are two choices: to either keep IT on premises the old way or go to the public cloud. There is a third way. In this episode, HPE';s Chris Eidler busts the myth by giving listeners a deep dive on how the just-announced HPE GreenLake Lighthouse cloud platform brings cloud-native IT to apps and data wherever they are. A first-of-its-kind hybrid cloud alternative.

Also available on:   Spotify    /   Apple Podcasts    /   Other podcast apps 

 

Transcript 

Robert:

 

Robert Christiansen:

Welcome to HPE's Tech Talk. I'm your host, Robert Christiansen. My guest today is my colleague and friend, Chris Eidler, and I couldn't be more excited about our conversation today. The theory behind HPE GreenLake Lighthouse has been a concept that, at its core, we've been working on for several years now and Chris has really helped bring it to life in a groundbreaking way. Thanks for being here, Chris.

Chris Eidler:

Oh, it's my pleasure, Robert. Happy to join you.

Robert:

Thank you so much, my friend. This has been quite a ride, hasn't it?

Chris:

You know, it really has been, but an exciting ride.

Robert:

It sure is. You know, Antonio just announced the HPE GreenLake Lighthouse at the HPE Discover. This is a solution you and I are both passionate about, and it's been a long time in the making. And before we even get into what it is, can you tell the listeners why HPE GreenLake Lighthouse is so monumental for them and for all of the global customers of HPE?

Chris:

You know, I think the best way to answer that is that the HPE GreenLake Lighthouse is a new way to consume the great infrastructure hardware and software that we develop here at the company. What we're doing with the HPE GreenLake lighthouse is taking that ... I almost like to think of it as a greatest hits playlist, Robert, and we turned it into a platform, right? So now you have an opportunity to discuss what your infrastructure and what your HPE partner can do for you in terms of enabling GreenLake services from us, instead of always having to talk about what's in the infrastructure. So it's a great way to transform the nature of the relationship and the interaction with us.

Robert:

Yeah, I remember when you and I were first discussing this, it's been almost two years now that... I was introducing you to the concept of the HPE GreenLake Lighthouse and how we were trying to articulate the need for on-premises cloud. But most people don't relate, you know, that there's a data center and there's infrastructure and there's a stack and there's all sorts of stuff that goes on whenever you consume like a public cloud, but they just don't see it. So getting people to make that leap, that there has to be a platform by which you deliver cloud on. And if you're going to deliver it on premises like we are doing with GreenLake, you got to have something to put it on. Wouldn’t you agree?

Chris:

Well, I totally agree. You know, and I think that not only is it clearly true, but I think there's a false image out there that when a customer decides how they want to deploy their IT needs, whatever those might be, that they either have to go totally old-school on prem or they have to go full on to public cloud. And both of those are great options, but I think that what we're doing with the HPE GreenLake Lighthouse platform together with HPE's GreenLake services that it hosts is let a customer, and let our clients, and really let the market know there's a place in between, right?

You can get a combination of the experience you expect from cloud with the infrastructure tuning, the infrastructure familiarity that we have with the enterprises needs on prem, and you can get them both all at once.

Robert:

And, I think this is seriously groundbreaking because on premises is different. There are just different things that we have to worry about, right? And so public cloud is ephemeral. It's highly scalable., Logically, you can go as much as you want to consume, right, and it doesn't seem to be any restrictions. But when you come on prem, especially client's backyard, you have to be mindful of such things like, you know, do they have a footprint space for it? Do they have power and cooling necessary? And how are they going to consume it? So let's break down Lighthouse a little bit if you would for us, Chris, and what are the characteristics of HPE GreenLake's Lighthouse platform today, we've announced?

Chris:

Well, what we've just announced at our Discover Conference was essentially the immediate availability of HPE GreenLake Lighthouse Mainstream Solutions Edition, MS. And this is the stack that we believe is able to host a really, really wide variety of workloads. It supports containers as a service from HPE GreenLake. It supports BM as a service. so think of it as the infrastructure as a service platform. The containers as a service are based on our Ezmeral technology, which is nice, and it's deeply integrated.

And so we now have an opportunity to pivot from discussing what's in the platform or discussing what that platform can do for you.

Now that being said, if you look inside the platform, we take advantage of ProLiant compute. We take advantage of Aruba networking. We take advantage of Nimble Storage. And the way that these are consumed by HPE GreenLake Lighthouse is as something that we call Lighthouse modules. And these modules are predefined, but highly configurable infrastructure elements or software elements that can get ingested by the HPE GreenLake Lighthouse. And then once the Lighthouse brings that in to the system, we can take those combinations of Nimble Storage. We can take those combinations of ProLiant compute. We can take the right combinations of networking configurations from the Aruba inside. And then we apply blueprints that allow us to orchestrate those systems into the configuration needed for the workload running on top of this. So it's a combination, to answer your question, Robert, of enterprise class kit, not designed to be an extension of the public cloud, but designed to be what our clients expect to run their businesses on for that 70 plus percent of their workloads that have an affinity to on-prem

Robert:

That's a really interesting piece.

If we were to unpack that just for a second, the public cloud is by its very nature extending into on-premises. That's basically what you see with Outpost or with Azure Stack or Anthos, et cetera, those type of things. What we are doing here with HPE GreenLake Lighthouse is to instantiate a cloud in our client, to give them those modules, those solutions in a way that are fully contained in and of itselves in that set that's coming into their operation.

And when you were describing that, that's what I constantly go back to is if you're going to have a private cloud, and you want it to be operationally similar in behavior and engagement with public cloud, there's a set of services that it has to do. And we've done that, but more importantly, I think that what you're driving here and what you are articulating is that it's configurable, but not customizable. Can you break that down a little bit about what that means? It's configurable, but not customizable.

Chris:

You know, that's a mantra here, configurable, not customizable, and back to that false choice that we talked about early on, I can either stay on prem the old way or go to the public cloud the new way. And this platform that we're talking about allows you to escape the choice of only having one or the other, but to stay on prem with what you're used to.

And the reason it has to be a platform is really, for two reasons, one is we need to be able to support rapid growth of a customer or of the adoption across the industry of HPE GreenLake. And in order to keep pace with what we anticipate is the demand for this platform, it has to be a platform and it has to be standardized. The other reason that it has to be a platform and a standardized platform at that is that we have to enable a rapid agility and velocity of service development in the ecosystem that runs on top of HPE GreenLake Lighthouses, including the services that are offered by GreenLake ourselves properly. So between those two variables is the need for a platform and a platform I would argue by definition needs to be standardized.

So now how do you not customize? And then the answer is it has to be very, very configurable to be able to handle the tailoring of what you need from a particular HPE GreenLake Lighthouse to the workloads you're running on it. So that configurability, in order to support enterprise price performance expectations became key to how we engineered the platform. Because as soon as it's customizable, you lose that agility benefit and you lose that scalability benefit.

And so what we've done in order to enable that configurability is invent some technology inside the metal orchestration layer within the Lighthouse. And we take the performance characteristics that HPE is legendary for within each component in our architecture and spanning every component in a north-south capability in support of the workload. And we tune the way we would tune a customer environment. And we then break it down into blueprints, blueprints and images. When we ingest Lighthouse modules, we then create a pool of these resources within the HPE GreenLake Lighthouse.

Then what happens is the client will engage with GreenLake um services layer and say, "I would like to run ... " Let's say it was Splunk, Robert, right? That's a great example of a workload that can benefit from our technology.

GreenLake's cloud software then makes an API call into the GreenLake Lighthouse saying, "I'd like to deploy Splunk," and we're going to respond, "We speak Splunk." [16:27]And once we get those parameters, we then apply those images. We apply those blueprints. So what we've done is standardized the platform, taken the best of what a bespoke platform would've done to performance optimize and figured out a way to sort of rehydrate that knowledge into the portion of the GreenLake Lighthouse that's hosting, in this case Splunk.

So bringing it all the way home for those of you that have had the stamina listen to that deep sort of dive into what we're doing is we can stay standardized so that the clients get the benefit of bullet train speeds. And yet, we're completely configurable to make sure that the particular outcome that a client's looking for can be obtained on the platform.

Robert:

That's great. Thank you. And one of the things that immediately gets asked, and I think we alluded to that just a little while ago when we were trying to compare, just mentioning the AWS Outposts, the Azure Stacks of the world, et cetera, you know, and the natural desire to drop the HPE GreenLake Lighthouse in amongst them as a peer, and I would immediately say, "Hey, well, that's great that you're comparing them to that, but let's just be clear. They're not in the same category." We're not trying to push out a public world into a private world. We're bringing a private cloud capability that gives, the client a set of services that's an experience that's very much like the public cloud, but on premises. And so when you hear that, how do you respond to that? And how do you look at that comparison to those other particular platforms?

Chris:

It's a legitimate question, you know.

Robert:

It is..

Chris:

You know, it's not a silly question. And the very first thing I always do is I make sure to not disrespect the work that's gone into those competitive platforms. They're amazing pieces of engineering and they deserve full credit. Happen to think ours are, too, right, but it's good work. It's the design center that's different, right? The design center, and I'm not talking about the technology design center necessarily, talking about the intent-based design center. The design center for a solution such as Outpost is extend a company's decision to standardize on AWS to enable some portion of that experience to happen on prem. So it extends the engineering thought processes of how's infrastructure engineered at Amazon?

So the good news is you get a ton of benefit if that's the experience you want to continue to have 'cause you're going to get more of that, okay. Our design center says, we want to take the experience that you've come to expect from your mission critical architecture on prem on which you've been running your business and that's 70% of the workloads that have proven that they have a stickiness to stay on or near prem. And we're instead extending your ability to continue to do that with a new platform. So are you trying to extend what you do today into a cloud experience? That's us. If you're trying to extend what you're doing on the cloud into your data center, then that's the approach from some of those competitors. So I-

Robert:

That's a great articulation of that, Chris.

Chris:

Yeah, I really think of it that way. I do think that clients run a lot of different cloud instances for a lot of different reasons. And we just happen to believe that there are a lot of reasons why running an on-prem cloud, whether on-premise in your data center, whether on-premise at a colo site, either operated by you or operated by someone else, or whether on prem as an Edge site, any of these examples we're just giving them an option to extend their data center, instead of extend their public cloud engagement.

Robert:

That's a great analogy. I think back on there was a manifesto that I wrote some almost two plus years ago where we were articulating this vision and I gathered up Kumar, who's our CTO and we read it and we worked it out and we were starting to develop this thinking around developing you know that core set of services that would go inside the HPE GreenLake Lighthouse and what those services had to be. And so this has come a long, long way, all right. So this wasn't just a knee-jerk reaction. I remember that when you and I first started having this conversations about you leading this group. I was just so pleased by your response. So what's the early response from the customers though? How's that going?

Chris:

Well, first, you know, thank you for the early insights into the need for a platform, Robert, 'cause we'd been running a team in another part of the company that was focused on enterprise solutions and performance, right? We were the team that built things out of the things we built. That's what we called ourselves. So in a way it was kind of like you were peanut butter and I was chocolate, and suddenly, we were a peanut butter cup. So it was a perfect match.

The early response from our customers is ... It's like they believe that this is the perfect piece at the perfect time for them. The answer would be enthusiastic support because we're at a point now where we can articulate the benefits of HPE GreenLake cloud services on HPE GreenLake Lighthouse without having to be defensive against the private cloud versus public cloud. And I think that we're coming along at a good time for our customers because they're understanding now. So are we, frankly, right? But they're understanding now that there's a right mix in here somewhere.

Robert:

There really is.

Chris:

And they're really, really hungry to find a vendor who understands the enterprise mission critical part of that right mix so well coming in with a platform that can support that part of their estate alongside the other platforms that they can choose from, from other vendors for other parts of their estate. So I think that the reaction from customers has been right solution, right time and we're really glad to see it coming from you, HPE.

Robert:

Well, Chris, thanks for joining me today.

Chris:

Oh, it was my pleasure. I mean you know we could talk about this for a long time. We're just getting warmed up.

Robert:

We could. 

And we definitely will. You and I will go deeper on the technology and services side the next time we talk, that's for sure.

Chris:

Perfect.

Robert:

All right, folks, it's been great looking back at this journey with the HPE GreenLake Lighthouse, and I know there's much more to come. Thank you for listening. I really hope you enjoyed this episode. Take a second to subscribe. Tell us what you think. Leave us a review. And if you want to hear more from HPE and this HPE Tech Talk, drop some topics in there. We'll bring in the experts. And I just want to say thank you for tuning in and have a great day. Bye-bye.

 


HPE Editor
Hewlett Packard Enterprise

twitter.com/hpe
linkedin.com/company/hewlett-packard-enterprise
hpe.com

HPE Editor
About the Author

HPE-Editor

Editor-in-chief for the HPE Advancing Life & Work blog.

Comments

Thanks for sharing!

Excellent blog! Love the clarity on configurable, not customizable.