You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think this is a great idea, especially for some combination of micro- and macrobenchmarks.
Most of what we have already in the suite is CPU-bound, and I wonder if some of this http benchmarks might be I/O bound. First, it would be good just to know that. But secondly, we may want to take care about how they roll up into aggregate numbers (probably through use of the aggregation-by-tagging that recently landed in pyperf).
Currently there's tornado_http but there's loads of high quality benchmarks for asyncio clients and servers
https://magic.io/blog/uvloop-blazing-fast-python-networking/
I also want to see if it's possible to make SelectorEventLoop perform faster than uvloop
The text was updated successfully, but these errors were encountered: