Ken Nalbone and Rich Stroffolino discuss all the announcements from HPE Discover, Nvidia bringing CUDA to ARM, Huawei slashing revenue, and the League of Entropy.
Stephen Foskett leads a discussion about how big of a change composable infrastructure is from the tried and true blade server.
News came out today in The Marker that Microsoft is looking into an acquisition of Mellanox. This would likely mark a major shift to in-house hardware development for Azure, and could form the basis of a composable cloud offering in Microsoft’s cloud. But it would also take a major (if un-heralded) player out of the market for everyone else.
I’ve been keen on rack-scale composable infrastructure for years. Decades even. But it’s only recently that we’ve had the technology to make it happen. You can now create a system or a rack that can flexibly allocate storage and compute using a shared I/O channel. But what if you could add more elements and “decompose” the server further? That’s what Liqid is promising with their latest announcements.
Openness, it seems, is “in the air.” While the Liqid folks were over at OCP, Russ White was over at the Open Networking Summit (ONS), where he attended the Linux Foundation (Networking), or LF(N), board meeting. What was interesting at this year’s summit is just how little the focus of open network is on bare metal routers and switches, and how much of the focus is on server-based and overlay networking. The last mile, as it were, in the data center is becoming ever more important. This just shows the question hanging over the network has not changed much in the last twenty years: how much intelligence will be pushed into the network, and how much in the host?
This week in Gestalt News:
– Stephen Foskett looks at Kasten brings enterprise-class data services into the cloud
– We talk to Packet CEO Zachary Smith for IT Origins
– Russ White looks at GPUs and Composable Computing
In our latest Tech Talk series with Liqid, Russ White looks at why composable infrastructure is ideally suited to answer a prickly question in GPU computing: How many processors is this job going to require?
In this edition of Gestalt News:
– Tom Hollingsworth takes a look at ExtraHop Reveal(x)
– DriveScale’s Tom Lyon sits down for an IT Origins interview
– Russ White considers if flexible scaleout in design could have mitigated the Meltdown vulnerability
Can the Gen-Z Consortium makes blade servers the future of the data center?
In response to a reader question on his look at Liqid’s composable infrastructure, Russ White frames an interesting question: is it easier to extend PCIe to support switching, and longer runs, or is it easier to design an entire protocol to (effectively) run PCIe over Ethernet? Liqid developed their solution based on former, but other composable infrastructure projects prefer an Ethernet based approach. It’s an interesting look into the benefits and drawbacks of both.