Slicker servers, heaving racks, NVMe invasion: It's been a big week in serverland

Too little too late

These guys are late to the party, and I don't see this concept working in the long haul.

"If you have racks of commodity server and are currently re-purposing servers between applications quite often, then having that process automated by products like Supermicro’s RSD could be seen as attractive."

^^ This is downright silly for 90% of the world's workloads. What everyone should be pursuing in a quest for effective utilization, is plain-old, boring compute clusters, where app (and other) folks can simply deploy a service, a process, a micro-service, a jvm, an instance of work, etc. Easy to manage (some of us have been doing this for 20 years), easy to maintain, and vastly cheaper than dedicating whole systems to a single app or a bunch of thick-provisioned VMs.