Google's First Tensor Processing Unit: Architecture

Re:Google's First Tensor Processing Unit: Architecture by KeplerBoy on Tuesday March 26, 2024 @05:33AM (#39825802)

Geohot is wrangling with unsupported consumer hardware.

The datacenter stuff is on a different architecture and driver stack. The number one supercomputer on the top500 list (frontier at ORNL) is based on AMD GPUs and AMD is probably more invested in supporting that.

[ Parent | on HN ]

Re:Google's First Tensor Processing Unit: Architecture by QuadmasterXLII on Tuesday March 26, 2024 @08:26AM (#39826992)

I mean, that's kinda nvidia's whole shtick: anyone can play around synthesizing cat pictures on their gaming GPU and if they make a breakthrough, the same software will transfer to X million dollar supercomputers.

[ Parent | on HN ]

Re:Google's First Tensor Processing Unit: Architecture by kkielhofner on Tuesday March 26, 2024 @09:44AM (#39827730)

I work with Frontier and ORNL/OLCF. They have had and continue to have issues with AMD/ROCm but yes, they do of course get excellent support from AMD. The entire team at OLCF is incredible as well (obviously) and they do amazing work.

Frontier certainly has some unique quirks but the documentation is online[0] and most of these quirks are inherent to the kinds of fundamental issues you'll see on any system in the space (SLURM, etc).

However, most of the issues are fundamentally ROCm and you'll run into them on any MIxxx anywhere. I run into them frequently with supported and unsupported consumer gear all the way up.

[0] - https://docs.olcf.ornl.gov/systems/frontier_user_guide.html

[ Parent | on HN ]