Repository navigation
docs(bench): republish the measurements on 5.7.1 - #1653
Merged
Merged
Conversation
Same Podman 5.7.0, podman-compose 1.5.0 and docker-compose 5.1.3 as the 3.4.1 run, so this one reads against it. The unchanged reference moved -6.5%, podup -5.7%: flat. Peak memory fell from a median of 8.0 MiB to 5.6 MiB per command; the logs rows fell -66% and -55%, the ps rows -28% and -22%. The config row prints +38% and a head-to-head on the same machine reads 11.7 ms against 11.6 ms; the page says so. Two things the method now states because they moved the numbers: the binary measured is the published musl asset (a plain cargo build on a glibc host links dynamically and reports twice the memory), and Podman's storage is counted and cleaned of test leftovers before the run (a podman ps went from 56 ms to 28 ms once a few hundred networks and images from integration tests were gone). bench/README.md's example now builds the musl target. Signed-off-by: Jaro-c <75870284+Jaro-c@users.noreply.github.com>
Jaro-c
added
type:docs
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
docs/benchmarks.md, the README summary anddocs/assets/bench.svgregenerated from a full run on 5.7.1: 1044 timed runs in ten minutes, none failed,bench/aggregate.pyandbench/chart.pyuntouched.What the numbers say
docker-compose) moved -6.5%, podup -5.7%: flat across nine releases.logsrows -66% and -55%,psrows -28% and -22%.config-heavy configprints +38% against the old table and a head-to-head on this machine, alternating binaries, reads 11.7 ms (3.4.1) against 11.6 ms (5.7.1). Published as measured, with that reading beside it.Two things that moved the numbers before the real run
target/release/podupfromcargo build --releaseon this glibc host: dynamically linked, 6.4 MB RSS on--versionwhere every published asset reads 3.3-3.7 MB. Discarded; the run above measures the publishedpodup-linux-x86_64asset.bench/README.mdnow says to build the musl target.podman pstook 56 ms and 28 ms after removing them. podman-compose, which shells out per call, moved +33% in that first run with no change of its own. The counts at the start of the real run are in the methodology paragraph.Signed-off-by: Jaro-c 75870284+Jaro-c@users.noreply.github.com