Evaluate using Profile-Guided Optimization (PGO) and Post-Link Optimization (PLO)

[zamazan4ik]

Hi!

Recently I checked many optimizations like PGO and PLO (mostly with LLVM BOLT) improvements on multiple projects including many databases - the results are available here. According to the tests, these optimizations can help to achieve better CPU performance in many cases. I think trying to optimize Fluvio's performance with PGO and PLO can be an interesting idea.

I can suggest the following action points:

• Perform PGO benchmarks on the project. If it shows improvements - add a note to the documentation about possible improvements in Fluvio's performance with PGO.
• Providing an easier way (e.g. a build option) to the build scripts to support building with PGO can be helpful for the end-users and maintainers since they will be able to optimize Fluvio according to their workloads.
• Optimize prebuilt binaries with PGO and PLO
• Testing Post-Link Optimization techniques (like LLVM BOLT) would be interesting too (Clang and Rustc already use BOLT as an addition to PGO) but I recommend starting from the usual PGO.

Here are some examples of how PGO optimization is integrated into other projects:

• Rustc: a CI script for the multi-stage build
• GCC:
  • Official docs, section "Building with profile feedback"
• Clang: Docs
• Python:
  • CPython: README
  • Pyston: README
• Go: Bash script
• V8: Bazel flag
• ChakraCore: Scripts
• Chromium: Script
• Firefox: Docs
  • Thunderbird has PGO support too
• PHP - Makefile command and old Centminmod scripts
• MySQL: CMake script
• YugabyteDB: GitHub commit
• FoundationDB: Script
• Zstd: Makefile
• Pydantic-core: GitHub PR
• file.d: GitHub PR
• OceanBase: CMake flag

I would be happy to answer your questions about all the optimizations above.

Thank you.

P.S. By the way, why the #3184 issue was closed? I guess it was just a simple auto-mistake. Could you please reopen it or move to the right place?