Improve benchmark performance #200
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sending 100,000 messages in parallel isn't a good way to use the system.
Before:
echo_bench.exe: [INFO] rate = 23975.514051
echo_bench.exe: [INFO] rate = 23774.676839
echo_bench.exe: [INFO] rate = 23807.165267
After:
echo_bench.exe: [INFO] rate = 28369.413333
echo_bench.exe: [INFO] rate = 28242.881411
echo_bench.exe: [INFO] rate = 28636.375119
By default, capnproto allocates 8k buffers. However, most capnp-rpc
messages are much smaller than this. This commit chooses more sensible
sizes (for the internal messages used by the benchmark test), and also
exposes the message size to users of the library for requests and
responses.
Before:
echo_bench.exe: [INFO] rate = 28369.413333
echo_bench.exe: [INFO] rate = 28242.881411
echo_bench.exe: [INFO] rate = 28636.375119
After:
echo_bench.exe: [INFO] rate = 40065.965541
echo_bench.exe: [INFO] rate = 40835.360431
echo_bench.exe: [INFO] rate = 40146.473277
TLS now generates a lot of distracting messages at debug level.
talex5
added a commit
to talex5/opam-repository
that referenced
this pull request
May 14, 2020
…pc and capnp-rpc-unix (0.7.0) CHANGES: - Update for x509 0.11.0 API changes (@talex5, mirage/capnp-rpc#196). - Update to new mirage network API (@Cjen1, mirage/capnp-rpc#198). - Add echo benchmark test (@Cjen1, mirage/capnp-rpc#197). - Estimate message sizes to improve performance (@talex5, mirage/capnp-rpc#200). By default, capnproto allocates 8k buffers, but most capnp-rpc messages are much smaller than this. Logging: - Fix application logging, to use library's log (@Cjen1, mirage/capnp-rpc#195). - Expose the endpoint logger (@Cjen1, mirage/capnp-rpc#195). - Only enable debug logging for capnp libraries in the calc example. TLS now generates a lot of messages at debug level (@talex5, mirage/capnp-rpc#200).
This was referenced May 14, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This tunes the benchmark to get better performance:
Results on my machine
Original:
Limit concurrency:
Reduce message sizes: