duke8253 edited a comment on pull request #7405: URL: https://github.com/apache/trafficserver/pull/7405#issuecomment-754894524
Running h2load with command `h2load --warm-up-time=10 -D 20 -c 4 -t 4 -m 4 https://127.0.0.1:9090/0`, this is what I get before and after the change, there are some improvements, but the difference is very minor. Both with `CONFIG proxy.config.exec_thread.limit INT 16`. However I'm not sure how the `tls-perf` is used in the original test mentioned in #7373. Before: ``` finished in 30.00s, 79022.75 req/s, 1.13MB/s requests: 1580455 total, 1580455 started, 1580455 done, 1580455 succeeded, 0 failed, 0 errored, 0 timeout status codes: 1580455 2xx, 0 3xx, 0 4xx, 0 5xx traffic: 22.61MB (23707065) total, 13.62MB (14281336) headers (space savings 94.21%), 0B (0) data min max mean sd +/- sd time for request: 59us 12.75ms 194us 86us 89.73% time for connect: 0us 0us 0us 0us 0.00% time to 1st byte: 0us 0us 0us 0us 0.00% req/s : 19634.19 19896.22 19755.60 134.63 75.00% ``` After ``` finished in 30.00s, 79617.90 req/s, 1.14MB/s requests: 1592358 total, 1592358 started, 1592358 done, 1592358 succeeded, 0 failed, 0 errored, 0 timeout status codes: 1592358 2xx, 0 3xx, 0 4xx, 0 5xx traffic: 22.78MB (23885610) total, 13.70MB (14360590) headers (space savings 94.21%), 0B (0) data min max mean sd +/- sd time for request: 60us 10.24ms 192us 74us 87.50% time for connect: 0us 0us 0us 0us 0.00% time to 1st byte: 0us 0us 0us 0us 0.00% req/s : 19801.47 20063.76 19904.39 111.93 75.00% ``` ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
