bachelor-thesis/benchmarks/benchmark-plan.md

peak-perf

• meassure ddr to ddr
• meassure ddr to hbm

All for 1KiB, 4KiB, 1MiB, 1GiB

All for HW and also SW path

All for intra-node, inter-node and inter-socket

--> conclude how much overhead DSA engine has

--> conclude size after which using HW makes sense this point is reached when submit overhead for hw execution is smaller than entire copy time for sw execution

submit // done

• single submit-and-wait
• multi submit
• batch submit

All with both 1 and 4 engines per WQ

All for 1KiB, 4KiB, 1MiB, 1GiB

All only on DDR and intra-node

--> conclude which work submission strategy is best for which size

--> conclude whether multiple engines significantly improve batch perf

mtsubmit // done

• multiple threads submit to the same WQ
• use 1,2,4,8,12 threads

All using DDR and 1MiB

All for 1 vs 4 engines

All on DDR and intra-node

--> conclude how bad mt submit hurts performance

--> conclude whether multiple engines help mt submit

cross-copy // done

• compare which is faster: xcopy, copy from source node, copy from dst node

All for both inter-node and inter-socket copy using DDR and 1MiB on 4E

--> conclude where a copy thread should live