chore(bench): add throughput for integer benchmarks #1741
Conversation
| /// Generate a number of threads to use to saturate current machine for throughput measurements. | ||
| pub fn throughput_num_threads() -> u64 { | ||
| let num_threads = rayon::current_num_threads() as f64; | ||
| // Add 20% more to maximum threads available. | ||
| (num_threads + (num_threads * 0.2)) as u64 | ||
| } |
There was a problem hiding this comment.
unclear this is enough, check with @tmontaigu probably
There was a problem hiding this comment.
Yes its unclear to know if its enough, for some ops it might for others not, or it could be enough for all,
I think a manual check needs to be done by running and trying different values and see if the throughput drastically changes
| bench_id = format!("{bench_name}::throughput::{param_name}::{bit_size}_bits"); | ||
| let elements = throughput_num_threads(); | ||
| bench_group.throughput(Throughput::Elements(elements)); | ||
| bench_group.bench_function(&bench_id, |b| { |
There was a problem hiding this comment.
unclear to me if this is the right way to call the throughput, did you confirm it works as expected ?
There was a problem hiding this comment.
Yes it works as expected, at least on my laptop. We'll need to double check on target machine to see if results are coherent.
|
We'll need also to add throughput benches for the GPU 🙂 |
Yes for sure, I just wanted to check for the design before deploying to all the functions. |
soonum
force-pushed
the
dt/bench/throughput
branch
from
e16e88c
to
f20e6e7
Compare
No description provided.