grpc: upgrade to 1.66.2 and use Codec v2

[vmg]

Description

This PR upgrades our GRPC version to 1.66.2 and introduces the new Codec v2 for serializing and deserializing protobuf messages. The goal of this codec is to enable memory pooling and hopefully increase real-world performance of Vitess clusters by reducing the number of memory allocations used when working with GRPC. You can can read up on the implementation details and next steps for the codec in this discussion PR.

This is the very first iteration of these changes, which only implements buffer pooling, but does not attempt to pool the objects themselves. Our hope is that this should already show in the benchmarks, so if it doesn't it'll be a bit disappointing! 🤞

Related Issue(s)

• codec: implement codec v2 planetscale/vtprotobuf#138

Checklist

• "Backport to:" labels have been added if this change should be back-ported to release branches
• If this change is to be back-ported to previous releases, a justification is included in the PR description
• Tests were added or are not required
• Did the new or modified tests pass consistently locally and on CI?
• Documentation was added or is not required

Deployment Notes

[vitess-bot]

Review Checklist

Hello reviewers! 👋 Please follow this checklist when reviewing this Pull Request.

General

• Ensure that the Pull Request has a descriptive title.
• Ensure there is a link to an issue (except for internal cleanup and flaky test fixes), new features should have an RFC that documents use cases and test cases.

Tests

• Bug fixes should have at least one unit or end-to-end test, enhancement and new features should have a sufficient number of tests.

Documentation

• Apply the release notes (needs details) label if users need to know about this change.
• New features should be documented.
• There should be some code comments as to why things are implemented the way they are.
• There should be a comment at the top of each new or modified test to explain what the test does.

New flags

• Is this flag really necessary?
• Flag names must be clear and intuitive, use dashes (-), and have a clear help text.

If a workflow is added or modified:

• Each item in Jobs should be named in order to mark it as required.
• If the workflow needs to be marked as required, the maintainer team must be notified.

Backward compatibility

• Protobuf changes should be wire-compatible.
• Changes to _vt tables and RPCs need to be backward compatible.
• RPC changes should be compatible with vitess-operator
• If a flag is removed, then it should also be removed from vitess-operator and arewefastyet, if used there.
• vtctl command output order should be stable and awk-able.

[vitess-bot]

Hello! 👋

This Pull Request is now handled by arewefastyet. The current HEAD and future commits will be benchmarked.

You can find the performance comparison on the arewefastyet website.

[codecov]

Codecov Report

Attention: Patch coverage is 87.50000% with 3 lines in your changes missing coverage. Please review.

Project coverage is 69.37%. Comparing base (da49d08) to head (fedf867).
Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
go/vt/servenv/grpc_codec.go	87.50%	3 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #16790      +/-   ##
==========================================
+ Coverage   69.35%   69.37%   +0.02%     
==========================================
  Files        1571     1571              
  Lines      204183   204188       +5     
==========================================
+ Hits       141601   141658      +57     
+ Misses      62582    62530      -52     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[vmg]

Wow so this actually worked:

https://benchmark.vitess.io/compare?old=main&new=fedf86784d6510a1837aa4d0f96817edab0b1fb8

We're missing TPCC which is not running atm for an unknown reasons but these numbers are huge. 3% increase in QPS and 13% reduction in allocated memory per requests is a huge win at this point in the lifetime of the project. And this is without adapting the vtprotobuf codegen to be aware of the new interfaces.

I think this is good to merge, as it's a drop-in replacement and the numbers are fantastic. The next steps involve updating vtprotobuf. I'll sit down with @mattlord or whoever has the cycles to pick this up to discuss the changes. 👌

[mattlord]

❤️ 🥳