Leoneazzurro5
Regular
Trento + MI200?
AMD CDNA Discussion Thread
- Thread starter Frenetic Pony
- Start date
Trento + MI200?
Bondrewd
Veteran
Yea.
DegustatoR
Legend
Bondrewd
Veteran
Yes, that lidded 3x reticle CoWoS option for HPC was made for MI200s.
Leoneazzurro5
Regular
N6 process.
Bondrewd
Veteran
All AMD parts are N6 now
Bondrewd
Veteran
I think it's ASE/SPIL's FOEB.
AMD is a big-big ASE house.
https://www.3dincites.com/2020/07/iftle-456-spil-fan-out-embedded-bridge-foeb-technology/
DavidGraham
Veteran
This isn't a one single wholesome GPU though, It's a two die GPU and appears as two GPUs to the system, juggling data between the dies is way slower than a single monolithic die, as the link between the two dies can only handle 200GB/s between each die, so aggregate performance numbers are theoretical numbers and should not be taken at face value.
Also, on contrast to NVIDIA's A100, this isn't primarily a matrix/AI GPU, but rather is a vector GPU primarily. As a result, AI enhancements are very modest compared to A100 even when combining the two dies together .. however, vector throughput is through the roof. AMD and NVIDIA is diverging here, AMD is still not fighting NVIDIA in the AI market, rather focusing solely on the HPC market.
Power consumption is 560W and is liquid cooled, which is curious considering the 6nm process.
Also, on contrast to NVIDIA's A100, this isn't primarily a matrix/AI GPU, but rather is a vector GPU primarily. As a result, AI enhancements are very modest compared to A100 even when combining the two dies together .. however, vector throughput is through the roof. AMD and NVIDIA is diverging here, AMD is still not fighting NVIDIA in the AI market, rather focusing solely on the HPC market.
Power consumption is 560W and is liquid cooled, which is curious considering the 6nm process.
Bondrewd
Veteran
That's all DC GPUs going forward due to NUMA.
A100 programming in some cases is already that due to wonky as balls split L2.
More, like 400.
No, this thing is just purpose-built.
They'll make a blingee matrix engine when the wider market wants their stuff.
500, 560 is the actual Frontier SKU as per stuffed into that Shasta blade.
https://www.amd.com/en/graphics/server-accelerators-benchmarks
AMD benchmarks.
On average MI250 is up to 2.3x faster in comparison with A100 in DP workloads with 1 GPU config mostly (except for LAMMPS).
The largest gains are in LINPACK of course.
A100 still has an edge in GPU to GPU communication bandwidth, so the difference with real multi-GPU configs will likely be smaller.
AMD benchmarks.
On average MI250 is up to 2.3x faster in comparison with A100 in DP workloads with 1 GPU config mostly (except for LAMMPS).
The largest gains are in LINPACK of course.
A100 still has an edge in GPU to GPU communication bandwidth, so the difference with real multi-GPU configs will likely be smaller.
D
Deleted member 2197
Guest
D
Deleted member 90741
Guest
MI200 series is old.
Cool folks are talking about MI300 already
Probably for delivery to El Cap
Cool folks are talking about MI300 already
Probably for delivery to El Cap
Bondrewd
Veteran
Yessss.
Obviously, big DOE heads quoted the El Cap timeframe already.
It's essentially 5Q after Frontier.
D
Deleted member 90741
Guest
I wonder if they will revise the perf target given latest exascale showdown with far east contenders.
Already saw some govt. head honchos on twitter debating.
Bondrewd
Veteran
El Cap is like puny 33MW so throwing moar nodes at the problem is totally feasible.
It's a closed off literally simulate nukes system so do they really have to?