On 12/12/2023 12:40, Pedro Tammela wrote:
[...] Hmm, that's interesting. So, IIUC, this implies that performance improvements have to have a certain magnitude to be useful for saving power, right? I.e., saving a few % of CPU usage on one core is not enough, but if the improvement is enough that you can move the workload to fewer cores, it will help because you can bring some cores offline/to idle. Or am I misunderstanding what you mean?
Yes exactly! Fewer cores also means fewer thermal pressure which also means FANs spinning slower :) Or potentially a longer server lifetime/cheaper server upgrade.
But when given more CPU room, applications might actually do more work! Take for instance TLS offload + zero copy, the CPU will only be really freed if the link/network stack is saturated.
I believe there are two approaches here to networking:
- Power saving vs Power efficient
I just remembered about QAT case on the Intel processors. It's a 12W coprocessor in the CPU die that beats a 56-core Sapphire Rapid on compression/decompression (TDP 350W). That would be the case where an optimization is so noticeable that the _power savings_ are measurable on the wall meter.