News for hackers. Stuff that matters.

A comment on Cerebras Wafer Scale Engine Gen2 7nm 2.6T Transistors.

Re:Cerebras Wafer Scale Engine Gen2 7nm 2.6T Transistors by blueblisters on Wednesday August 19, 2020 @04:35AM (#24208374)
Might it not be more optimal to attach the individual GPU dielets onto a silicon interposer to avoid the yield issues that are inherent when making a large monolithic chip, and save some of the overheads of dealing with dead dielets? I think I read about some researchers/companies going that route but haven't kept up.
[ Parent | on HN ]
Re:Cerebras Wafer Scale Engine Gen2 7nm 2.6T Transistors by lsb on Wednesday August 19, 2020 @04:57AM (#24208479)
Part of the genius of the Cerebras on-chip routing is that it can route around malfunctioning dielets.
[ Parent | on HN ]
Re:Cerebras Wafer Scale Engine Gen2 7nm 2.6T Transistors by gridlockd on Wednesday August 19, 2020 @05:32AM (#24208656)
Even then, wouldn't it be better to just use commodity chips on an interposer?

I don't actually know, but that seems to be the route the industry is going and economics of scale will likely make it the winning approach.

[ Parent | on HN ]