Intel Xeon 6980P vs. AMD EPYC Power Efficiency / Performance-Per-Watt Benchmarks

[jochendemuth]

Because it is clear that you see the servers only in pictures.

Speed of execution of tasks is not important? How many simultaneous users can be served? How much space does it occupy???

The best in perf/power is EPYC 8534PN (64 Cores, TDP 175W). In terms of efficiency, it destroys EPYC 9684X, EPYC 9654, EPYC 9754 etc.

Why is it not so popular?

Because to perform the same task in time, to handle as many users as on faster processors, you need to buy 50% more of them.

... and why wasn't it even part of this review?

[geerge]

I still feel my statement stands. They are flawed direct comparisons to make all around on an architecture basis.



Regarding power usage, with some of the workloads, I'd take 62% more power for 100+% performance gain. Overall it really depends on what your workload is to measure what is worth it. Some workload really benefit the Xeons and others benefit the Epycs while some are dead even.

9754 and 6980P are about as similar as products from different companies can be. AMD's little cores are much closer to their big cores compared to intel, intel's little cores can't do AVX512 and are much weaker. What would be a better AMD CPU to compare to the 6980P? There's a doubled cache disparity regardless of which of 9754 and 9864X is chosen, comparing CPU's of different core counts is objectively flawed, intel has some tech that no AMD part has (AMX, accelerators, memory bandwidth advantage). Maybe if we're trying to compare arch not CPU the intel part should be TDP-limited to the 9754's 360W (Michael please do this), other than that I don't think there is a more ideal part to compare with.

Sure, for a specific workload intel may be the clear winner, namely the 6980P has the memory bandwidth advantage, the AMX advantage and the accelerator advantage which counts for something. But odds are people will be taking the 62% extra power hit for ~23% more performance (which might be worth it, might not).