All in Composable Infrastructure
Sandesh and I went to the HPC + AI Wall Street conference where friend of the show Ryan Quick was a featured speaker. There were a bunch of sessions focusing on everything from Quantum computing and Neuromorphic computing to Artificial Intelligence and Big Data. The show was run by Tabor communications that is the publisher of HPC Wire, Datanami, and EnterpriseAI to name a few.
I have been a fan of disaggregation and composable infrastructure for some time now. It is a great way to get better utilization out of your infrastructure, and improve flexibility and the programmability of your resources. Even companies like HPE have validated the market with their Synergy platform, but, adoption has been slow. Today we are going to be talking with a company that is looking to change that.
Today we are going to be talking with a company that is re-thinking the way computers work. As workloads become increasingly more resource intensive, this company is leading the Big Memory and Memory Centric Computing charge.
So I have talked in the past about the rise of disaggregation and compassable infrastructure, and a lot of that was on display at VMWorld. I first ran into Liqid at the Open Compute Platform summit a few years back. They drew a lot of attention for showing how you could provision hardware assets as easily as you could provision VMs. Because it is all REST based, you could add physical hardware provisioning into your DevOps process. Basically, all of the components are joined together over a PCIe fabric to allow the dynamic provisioning of components.
