In order to compare upcoming changes for their effectivness, measure
performance by counting opertions and the runtime of each operation over the time. Accumulate all tests in a single instance, so make it complicated over the time. If you wait long enough, you will notice the expiry of old flows.
Details
Details
- Reviewers
kp - Group Reviewers
network - Commits
- rG33c1bdfc3e09: tests/libalias: Add perfomance test utility
Run the test for 15 seconds, while applying multiple rounds of:
- try to generate 15000 valid connections
- let 2000 random external sources ask for random ports
- let a single random external source hammer with 1500 identical packets
- try to respond to the valid connections in a random order
$ perf 15 15000 2000 1500 RND SEC NAT RND ATT UNA 1 0 13 12 12 12 2 0 13 12 11 12 3 1 14 13 12 12 4 1 17 13 12 12 5 2 18 15 12 12 6 2 23 15 12 12 7 3 23 17 12 11 8 3 28 17 12 12 9 4 33 18 12 12 10 5 35 19 12 12 11 6 39 22 12 12 12 6 42 21 12 11 13 7 44 21 13 12 14 8 48 24 13 12 15 9 52 25 13 12 16 10 57 25 13 12 17 11 61 26 13 12 18 13 63 27 13 12 19 14 68 29 13 Results Rounds : 18 NAT ok : 285000 NAT fail: 0 UNNAT ok : 270749 UNNAT fail: 0 RANDOM ok : 0 RANDOM fail: 38000 ATTACK ok : 0 ATTACK fail: 28500 ------------------- Total: 622249
Diff Detail
Diff Detail
- Repository
- rS FreeBSD src repository - subversion
- Lint
Lint Passed - Unit
No Test Coverage - Build Status
Buildable 39344 Build 36233: arc lint + arc unit