News for hackers. Stuff that matters.

A comment on Google's First Tensor Processing Unit: Architecture. Parent comment

Re:Google's First Tensor Processing Unit: Architecture by dagmx on Tuesday March 26, 2024 @02:40AM (#39824830)
Don’t underestimate CUDA as the moat. It’s been a decade of sheer dominance with multiple attempts to loosen its grip that haven’t been super fruitful.

I’ll also add that their second moat is Mellanox. They have state of the art interconnect and networking that puts them ahead of the competition that are currently focusing just on the single unit.

[ Parent | on HN ]
Re:Google's First Tensor Processing Unit: Architecture by latchkey on Tuesday March 26, 2024 @04:15PM (#39832343)
This moat is going to get paralleled over the next few years. First off Mellanox is unobtanium with 52+ week lead times.

GigaIO has a PCIe fabric solution that is a fraction of the cost of Mellanox and available today. This enables up to 64 GPUs to appear on a single system.

We're also seeing the ultraethernet stuff come online as well, but that'll have to wait for PCIe6.

[ Parent | on HN ]