Switch to Snappy compression for metadata

[joshtriplett]

No description provided.

[rust-highfive]

r? @Mark-Simulacrum

(rust_highfive has picked a reviewer for you, use r? to override)

[joshtriplett]

Inspired by #75732 , but easier to ship, and should provide even more performance. Passes tests locally. Let's see how it affects performance.

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion

[bors]

⌛ Trying commit 574f6be with merge 50541435b8c0b771394cb492b0b2fbdbc3c3ddd9...

[bors]

☀️ Try build successful - checks-actions, checks-azure
Build commit: 50541435b8c0b771394cb492b0b2fbdbc3c3ddd9 (50541435b8c0b771394cb492b0b2fbdbc3c3ddd9)

[rust-timer]

Queued 50541435b8c0b771394cb492b0b2fbdbc3c3ddd9 with parent e15510c, future comparison URL.

[rust-timer]

Finished benchmarking try commit (50541435b8c0b771394cb492b0b2fbdbc3c3ddd9): comparison url.

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. Please note that if the perf results are neutral, you should likely undo the rollup=never given below by specifying rollup- to bors.

Importantly, though, if the results of this run are non-neutral do not roll this PR up -- it will mask other regressions or improvements in the roll up.

@bors rollup=never

[alex]

I don't know this code well, so I'm not sure what compat/iteroperability needs are, but zstd is another compression algorithm that might be interesting to play with. It's a hair slower than snappy, but compresses much more, so the IO savings may be worth the CPU cost: https://facebook.github.io/zstd/

[Mark-Simulacrum]

@bors r+

Thanks!

[bors]

📌 Commit 574f6be has been approved by Mark-Simulacrum

[bors]

⌛ Testing commit 574f6be with merge 7fc048f...

[bors]

☀️ Test successful - checks-actions, checks-azure
Approved by: Mark-Simulacrum
Pushing 7fc048f to master...