Speedcopy #20
Merged
Speedcopy #20
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
Turns out that benchmarking from a live interface is silly, because it's always blocked on the interface, not on the actual code. Switched to benchmarking from a file, which produces much more repeatable results.
Used benchmark to test this out. Old: (/tmp/bench is tools/benchmark with the old code) gconnell@zombie-test1.mtv:/$ /tmp/bench -file /tmp/foo Took 3.334896s to process 1000000 packets, 3.334us per packet, 0 nil packets gconnell@zombie-test1.mtv:/$ /tmp/bench -file /tmp/foo Took 3.329675s to process 1000000 packets, 3.329us per packet, 0 nil packets New: (/tmp/bench2 is tools/benchmark with the new code) gconnell@zombie-test1.mtv:/$ /tmp/bench2 -file /tmp/foo Took 1.588695s to process 1000000 packets, 1.588us per packet, 0 nil packets gconnell@zombie-test1.mtv:/$ /tmp/bench2 -file /tmp/foo Took 1.580306s to process 1000000 packets, 1.58us per packet, 0 nil packets About a 50% reduction in the initial packet copy. This seemed to work pretty repeatably.
Conflicts: pcap.go
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Outdated suggestions cannot be applied.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.
Speed up the copy of packet data from C into Go, using C.GoBytes. Also avoid a copy of the packet header struct. Together, this seems to reduce packet read speed by ~50%
Added a benchmark binary for testing this change.
Old:
(/tmp/bench is tools/benchmark with the old code)
gconnell@zombie-test1.mtv:/$ /tmp/bench -file /tmp/foo
Took 3.334896s to process 1000000 packets, 3.334us per packet, 0 nil packets
gconnell@zombie-test1.mtv:/$ /tmp/bench -file /tmp/foo
Took 3.329675s to process 1000000 packets, 3.329us per packet, 0 nil packets
New:
(/tmp/bench2 is tools/benchmark with the new code)
gconnell@zombie-test1.mtv:/$ /tmp/bench2 -file /tmp/foo
Took 1.588695s to process 1000000 packets, 1.588us per packet, 0 nil packets
gconnell@zombie-test1.mtv:/$ /tmp/bench2 -file /tmp/foo
Took 1.580306s to process 1000000 packets, 1.58us per packet, 0 nil packets
About a 50% reduction in the initial packet copy. This seemed to work pretty
repeatably.