While a lot of people focus on the floating point and integer processing architectures of various kinds of compute engines, we are spending more and more of our time looking at memory hierarchies and interconnect hierarchies. …
We like datacenter compute engines here at The Next Platform, but as the name implies, what we really like are platforms – how compute, storage, networking, and systems software are brought together to create a platform on which to build applications. …
Not everybody can afford an Nvidia DGX AI server loaded up with the latest “Hopper” H100 GPU accelerators or even one of its many clones available from the OEMs and ODMs of the world. …
You can’t be certain about a lot of things in the world these days, but one thing you can count on is the voracious appetite for parallel compute, high bandwidth memory, and high bandwidth networking for AI training workloads. …
Imagine, if you will, that Nvidia had launched its forthcoming “Grace” Arm server CPU three years ago instead of early next year. …
There are a lot of things that compute engine makers have to do if they want to compete in the datacenter, but perhaps the most important thing is to be consistent. …
Nvidia got a little taste of hardware, and the company’s top brass have decided that they like having a lot of iron in their financial diet. …
It is difficult not to be impatient for the technologies of the future, which is one reason that this publication is called The Next Platform. …
The golden grail of deep learning has two handles. On the one hand, developing and scaling systems that can train ever-growing model sizes is one concern. …
All Content Copyright The Next Platform