The following are the benchmark results for parallel speedup of rayon, tokio, and async-std on a 16-thread machine. (Core i7-12700K big core only, Gentoo Linux)
- The input variable is the "Available parallelism", e.g. how many parallel work units are spawned at the same time.
- The measured time is from the start of the work unit spawning to the end of the last work unit.
- Before and after each iteration, the threadpool is always empty.
- Hence the low parallelism results should be intepreted as "the runtime latency of a burst workload"
- "light": 10 random uncached memory read.