The Six Five team discusses NVIDIA advancing instant AI and what that means.
Watch the clip below:
If you are interested in watching the full episode you can check it out here.
Disclaimer: The Six Five Webcast is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors and we do not ask that you treat us as such.
Daniel Newman: Let’s talk about NVIDIA, Pat, so give the latest one over to you but their Base Command goes GA.
Patrick Moorhead: Think of Base Command as a way to get into a DGX SuperPOD without you having to roll it in to your own premises and not having to wait for, let’s say, one of the Tier 1 CSPs who likely won’t take the exact architecture of the DGX SuperPOD. So essentially it is a managed offering right now through Equinix and this sucker went GA. They have partners here. They have NetApp as a partner. I already talked about Equinix, and also Weights & Biases who, I guarantee you, you’ve probably never heard of, but they’re a leading provider of machine learning development tools out there for ML ops.
Because it’s one thing just to stick a GPU in a server and put it out there and call it a day, but with Base Command you have a full end-to-end develop an environment. And whether that’s for NLP autonomous vehicles, you can swipe a credit card… Actually, I don’t know if you can actually swipe a credit card, but you can go and get this service. And it is, to me, and NVIDIA is not going to say this, is this is an offering to keep multiple people in check.
On the OEM side who aren’t doing DGX SuperPODS, and on the cloud side, the AWS, the Azure and the GCP, this is a time to market play, but DGX is a real business. And I like that they’re doing as-a-service and, as you know, Daniel, a lot of the traditional on-prem folks like Dell are also partnering with Equinix for their as-a-service capabilities.
Daniel Newman: Yeah, I believe GreenLake also is doing a service through HPE so, yeah, Equinix is finding its footing in terms of taking advantage of co-location to offer private cloud at scale. Very interesting, of course. The whole announcement, Pat, you hit it. This isn’t one that probably takes a ton more effort. They did have some really good early customer wins. Adobe, for instance, is using it to accelerate their data science.
And look, all kinds of these advanced computing capabilities being put in the cloud just makes a lot of sense. We’re seeing the way AWS and Azure have taken Quantum and put it in the cloud. Simulation, you’re going to see this with HVC. How do you democratize advanced compute capabilities and make them available to people on a subscription basis and a consumption platform and manner? Makes a lot of sense. It’s going to bring a lot more enterprises into the fray. Once these enterprises understand the power, the capabilities, they either scale up and spend a lot of money on these services through consumption, or they become customers and they buy these machines and they actually take the whole service on and bring the whole thing in-house. Either way-
Patrick Moorhead: Well, they want a discount, too. I mean, it starts at $90,000 for a three-month commitment. That’s a deal for a DGX SuperPOD.
Daniel Newman: Yeah, we’ll go in and split it. We’ll figure out how to grow my hair back.