HPE Nimble dHCI - Day 2 Experience

9:19 · Watch on YouTube ↗

Transcript 1,043 words · about 7 min to read

Auto-generated captions from YouTube, not hand-corrected, so names and technical terms may be imperfect. The video is authoritative.

>> Okay welcome back to our series in which we're looking at, HPE's dHCI solution inside of the CTO Advisor Hybrid Infrastructure. Let's get down to, bare knuckle roots. Is this a viable infrastructure? Is this a viable design? Can you take a traditional architecture, where you have a separate storage, SAN and you have a separate compute platform, combine that together via software and automation and make management easier. This is the video in which we talk about, kind of day two, going into day three, management of dHCI, and the value that we see, or where it fell short in the CTO Advisor Hybrid Infrastructure.

All right so, we know the overall problem of inside of the CTO Advisor Hybrid Infrastructure. We have, ice cream inside of cake. We want more visibility, more capability, more flexibility, and dHCI promises that. We saw in the initial install, that the 15 minute installation claim, while a marketing claim, gave more truth to the ease of installation. Now let's talk about management. The second level problem that the CTO Advisor, specifically had, with being able to, upgrade from one version of vSphere to another version of vSphere from one control plane, to another version of the control plane.

What did we expect? We expected HPE to basically have figured out all of the automation needed to, enable, the increased technical capability. Lets talk industry wide for a second here. I don't care if you're talking about dHCI, HCI, public cloud, AI, ML infrastructure. You need automation, in order to do more with less. I cannot have my engineers, spending time learning the commands to deploy new VLANs. Learning the commands and the clicks, button clicks to deploy new LANs and volumes and server ProTech profiles, on a storage array.

I can't have them troubleshooting the minutia, of profile performance inside of a data center, that's as complex as even the CTO Advisor's simple deployment. So what did I expect from HPE? I expected the automation to simply work and did it simply work? Well, you know what, it's a mixed bag. This is the CTO Advisor. We kind of, remove the marketing fluff and tell you exactly what we found. We started out, with an older version of their solution, and then, they upgraded our demo unit to the latest version, Tom put it through its paces.

And what we discovered is that, the architecture, makes sense. The vision makes sense, but, not even the great HPE, who I've been a customer of for years, can think of every instance. That automation needs to be implemented and how, across a thousand different customers, how every nuance of the difference. And we ran into some of those challenges, but overall, HPE did deliver on the promise of putting automation on top of a, traditional stack and making our operations easier. Household, give you an example.

I talked about how Tom, gave me a quote. of 40 to 80 hours of looking for, the right version of the hardware storage controller. The right hard drives for VSAN the right Intel parts, for networking et cetera, to be validated as a HCI solution. HPE, basically put that into a repository for, the CTO advisor team to simply go to, select a pack, pull it down, deploy. And now Tom's 80 to 40 hours of work, to research, what versions of what VMware and hardware revisions, is reduced to basically a 15 minute check.

And this is the 15 minute claim come true, to deploy, the next version of HPE's pack. So how did that work in practice? So what does this all mean to the business? Like why should CIO, CTOs care? If we need to move faster, we need staff that, can basically move at the speed of the business. When Tom came to me and said that, it would take 40 to 80 hours to upgrade our VCI, our HTI platform, to the latest version of vSphere, I hit a pause.

My business, cannot take out the time , to dedicate one of my top engineers, if not my top engineer, to do a menial tasks, such as upgrade, from one version of a hypervisor to another version of a hypervisor. This service up and our ability to meet SLAs. When, I need to deploy a new ERP landscape, and I'm spending, 40 to 80 hours doing the simplest blocking and tackling things, to hardware and add on that hardware procurement, change management, et cetera. I'm just not meeting, the call to action, from the business.

Automation is needed in order to make the SLAs, that the business is driving CTOs and CIOs to meet possible. The question is, is HPE doing this with dHCI? Let's talk a little bit, from a technical perspective. InfoSight, I can't overemphasize this. I'm surprised. I've been critical of HPE and some other marketing claims. I'm surprised, InfoSight isn't pasted over all the branding and all the capabilities. By far InfoSight is the thing that changes, my operations and capabilities and my ability to deliver, applications faster.

To monitor those applications, so that I can have staff like Tom, doing things like learning Kubernetes or cloud native architectures. Or how to better secure my data, across public and private cloud. These things, these improvements HPE is making in their dHCI platform, are table stakes. They are needed in order to move the ball, of enterprise IT, further along the data analysis route. The ability to, get insights from our data, so that, we can build better applications and answer, the questions around, data from the Edge to the cloud better.

Speaking, about answering questions, from the Edge to the cloud better, the next video series, we're going to look deeper into this AI, ML capability. This ability to run applications on this dHCI stack and whether or not HPE is meeting those claims. com, is the website for the CTO Advisor business. And, I'm on Twitter, @CTOadvisor is the Twitter handle. DMs are open. You can challenge me, go back and forth. Was I too easy on HPE? Was I too hard on them?