As work happens on I/O pieces I've been building specialized micro-benchmarks (ex. gh-120754 Speed up open().read() pattern by reducing the number of system calls and others have gh-117151: IO performance improvement, increase io.DEFAULT_BUFFER_SIZE to 128k), it would be nice to have more general benchmarks to validate I/O performance for common cases.
Talking a little with people at PyConUS there was some interest in the tests, and a general desire for I/O tests not to be enabled by default, but to be a group which can be manually run.
General I/O shapes I'm hoping to add benchmarks for:
- read/write all of the byes of a file in a single call (including pathlib.Path.read_text, pathlib.Path.write_text)
- read/write many small files (ex.
.pyc files, maybe just compile_all?)
- streaming bytes read/write (ex. to a pipe / console such as stdin/stdout/stderr, non-seekable devices)
- read/write a zipfile, tarfile (read + seek, write + seek, in particular buffering behavior)
- use
zipimport
- Create a
zipapp
- Multi-threaded write to stdout, stderr (ex. logging in a large application/codebase)
Note: With these aiming to stay at the Binary / Bytes IO layer as much as possible (not touch Text I/O for now)
As work happens on I/O pieces I've been building specialized micro-benchmarks (ex. gh-120754 Speed up open().read() pattern by reducing the number of system calls and others have gh-117151: IO performance improvement, increase io.DEFAULT_BUFFER_SIZE to 128k), it would be nice to have more general benchmarks to validate I/O performance for common cases.
Talking a little with people at PyConUS there was some interest in the tests, and a general desire for I/O tests not to be enabled by default, but to be a group which can be manually run.
General I/O shapes I'm hoping to add benchmarks for:
.pycfiles, maybe justcompile_all?)zipimportzipappNote: With these aiming to stay at the Binary / Bytes IO layer as much as possible (not touch Text I/O for now)