3 ms·
Woah. The photos. IBM people are wearing t-shirts and jeans? It's not 1970 any more... Anyway. To the content: 7 nm? That's not cutting edge, but it's pr
by AnimalMuppet 5y ago
Woah. The photos. IBM people are wearing t-shirts and jeans?
It's not 1970 any more...
Anyway. To the content:
7 nm? That's not cutting edge, but it's pretty good. They do say that they have 2 nm in development.
> including the new 7 nm Telum chip that can facilitate on-device AI inferencing that’s 20 times faster than sending an AI request to an x86 server in the cloud
That's pretty underwhelming. An amazing amount of things should be 20x faster than sending a request to an x86 server in the cloud. I mean, AI in the mainframe is an interesting development. But in the mainframe means you don't have network delays compared to in the cloud. They only got 20x out of that? Plus, out of a custom chip vs an x86? Not impressive.
22 billion transistors on each AI chip is pretty good.
> An IBM zSystem requires 50% less energy than x86 systems to run the same workload. zSystems can run at 100% utilization, whereas x86 systems run at far lower utilizations.
Both are good points.