-
Notifications
You must be signed in to change notification settings - Fork 4
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add parallel compression #167
Conversation
But implementation is not yet concurrent
Also change to a null writer in benchmark, so we don't use so much memory
Benchmarks show it faster, nice
Really nice performance improvement! Thank you for the contribution! |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #167 +/- ##
==========================================
+ Coverage 77.34% 79.34% +2.00%
==========================================
Files 7 7
Lines 384 460 +76
==========================================
+ Hits 297 365 +68
- Misses 61 67 +6
- Partials 26 28 +2 ☔ View full report in Codecov by Sentry. |
Tried to add a bunch of silly tests to make codecov happy :) |
Thank you for your contribution! |
zstdseek is great, but slow. Let's use more cores!
FrameSource
abstraction, a function that returns a bunch of frames in orderConcurrentWriter
interface, which addsWriteMany
to add all frames from a FrameSource using concurrencyWithConcurrency
to manually set concurrency level, andWithWriteCallback
to run a callback each time a frame is writtenWrite
vsWriteMany
, and benchmarks themResults of benchmarks on my machine:
In tests of zstdseek, I see speedups of anywhere between 2x and 5x. I'm not sure why I can't quite hit the 12x I'd expect.