Skip to content

Add information about pbs performance - #1192

Open
jjhelmus wants to merge 1 commit into
mainfrom
jjh/benchmarks
Open

Add information about pbs performance#1192
jjhelmus wants to merge 1 commit into
mainfrom
jjh/benchmarks

Conversation

@jjhelmus

@jjhelmus jjhelmus commented Jul 24, 2026

Copy link
Copy Markdown
Contributor

Add a document (BENCHMARKS.md) which discusses the performance on python-build-standalone in pyperformance benchmarks compared to other CPython distributions.

@jjhelmus

Copy link
Copy Markdown
Contributor Author

Figure for BENCHMARKS.md

pbs-314-comparison

@jjhelmus
jjhelmus marked this pull request as ready for review July 24, 2026 13:15
@zsol
zsol self-requested a review July 27, 2026 11:01
Comment thread BENCHMARKS.md
src="https://github.com/user-attachments/assets/5499f9ce-ee02-4485-baa2-9d982fe457e6"
/>

python-build-standalone provides high-performance CPython distributions designed for compatibility across a broad range of Linux distributions, macOS releases, and Windows versions. It's builds incorporate compiler optimizations such as profile-guided optimization (PGO), link-time optimization (LTO), and, where appropriate, BOLT post-link binary optimization. Together, these techniques improve runtime performance while preserving portability.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit: I feel like this might be appropriate for a blog post, but when you're already in astral-sh/python-build-standalone/blob/main/BENCHMARKS.md this doesn't add too much value. At the very least it's out of place, sitting below the chart. I'd start immediately with the "The figure above" paragraph

Comment thread BENCHMARKS.md

### Figure details

The figure above compares CPython 3.14.6 performance across several distributions using pyperformance. Each violin represents the distribution of per-benchmark mean runtime ratios between an alternative CPython distribution and python-build-standalone on the same platform and architecture. Ratios greater than 1 indicate that python-build-standalone was faster; ratios less than 1 indicate that the alternative distribution was faster.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

maybe smuggle a link to https://en.wikipedia.org/wiki/Violin_plot ? I wasn't familiar with this until I looked it up

Comment thread BENCHMARKS.md
Comment on lines +19 to +28
From top to bottom, the distributions shown are:
* The Docker `python:3.14` image for x86-64, providing CPython 3.14.6.
* A conda-forge Python 3.14.6 environment for `linux-64`.
* The system Python 3.14.6 in a `fedora:44` x86-64 Docker container.
* The system Python 3.14.6 in a debian:forky x86-64 Docker container.
* CPython 3.14.6 from the Python.org macOS installer on an arm64 Mac.
* A conda-forge Python 3.14.6 environment for `osx-arm64`.
* CPython 3.14.6 installed through Homebrew on an arm64 Mac.
* CPython 3.14.6 from the Python.org Windows installer for x86-64.
* A conda-forge Python 3.14.6 environment for `win-64`.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should we include more precise versions that will allow people to reproduce the benchmark? e.g. the docker image hash, or the link to the python.org macos installer

Comment thread BENCHMARKS.md
Comment on lines +3 to +7
<img
width="1820" height="1730"
alt="Shows a chart with violin plots with benchmark results"
src="https://github.com/user-attachments/assets/5499f9ce-ee02-4485-baa2-9d982fe457e6"
/>

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should we include the raw data too, not just the image?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants