Ampere super computer coming this summer

MangoSeed

[H]ard|Gawd
Joined
Oct 15, 2014
Messages
1,447
Indiana University is the proud owner of the first operational Cray “Shasta” supercomputer on the planet. The $9.6 million system, known as Big Red 200 to commemorate the university’s 200th anniversary and its school colors, was designed to support both conventional HPC as well as AI workloads. The machine will also distinguish itself in another important way, being one of the world’s first supercomputers to employ Nvidia’s next-generation GPUs.

Those GPUs are expected to be plugged into Big Red 200 later this summer – that according to Brad Wheeler, vice president for information technology and chief information officer. The exact nature of those GPUs is unknown, which is understandable, inasmuch as Nvidia has not announced they are even on the way. The most likely explanation is that they will be the next-generation Tesla GPUs based on the upcoming “Ampere” architecture.

https://www.nextplatform.com/2020/0...e-first-production-cray-shasta-supercomputer/
Looking almost guaranteed now that nvidia will talk about Ampere’s HPC and AI chops in detail at GTC in March.
 
Please be super fast and not be more expensive than Turing!
That way GPUs in the $200-$300 range will get faster!
 
Please be super fast and not be more expensive than Turing!
That way GPUs in the $200-$300 range will get faster!

If they launch it the same way they did Pascal (new process node), the die sizes will be significantly smaller. This mean,one production/yields ramp-up, you should see lower prices for higher performance.

That Supercomputer part is 70% faster than Volta, so it probably has similar die size. I'm guesstimating smaller dies for the consumer parts, giving around 40-50% higher shader performance at each consumer prce-point (who knows how much they're increasing RT performance, but probably at least 2x).
 
I am far more interested in knowing if this is a Monolithic core or an MCM implementation.
 
Back
Top