YGGverse/aquatic
1
0
Fork 0
mirror of https://github.com/YGGverse/aquatic.git synced 2026-03-31 17:55:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

README: update aquatic_ws benchmark section

Browse source
This commit is contained in:
Joakim Frostegård 2021-08-18 19:59:20 +02:00
parent 482870b174
commit d6e7898fb2
1 changed files with 9 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

16
README.md
View file

@ -156,16 +156,18 @@ of the `aquatic_http` section above.
#### Benchmarks #### Benchmarks
Performance was compared to Performance was compared to
[wt-tracker](https://github.com/Novage/wt-tracker) and [bittorrent-tracker](https://github.com/webtorrent/bittorrent-tracker), using [wt-tracker](https://github.com/Novage/wt-tracker) and [bittorrent-tracker](https://github.com/webtorrent/bittorrent-tracker) using
`aquatic_ws_load_test` with one worker, 16 connections and 2 offers per request. Both comparison trackers reached 100% single-core utilization. `aquatic_ws_load_test` with one worker, 16 connections and 2 offers per request.
Server responses per second, best result in bold: | tracker | responses per second |
| ------------------ | -------------------- |
| aquatic |  __126k__ |
| wt-tracker |  54k |
| bittorrent-tracker |  28k |
| aquatic |  wt-tracker | bittorrent-tracker | Both reference trackers reached 100% single-core utilization.
| ----------- | ------------ | ------------------ |
| __126k__ | 54k | 28k |
This benchmark is not very realistic, as it simulates a tiny number of clients each sending a large number of requests, but I still think that it gives a useful indication of relative performance. This benchmark is not very realistic, as it simulates a tiny number of clients, each sending a large number of requests. Nontheless, I think that it gives a useful indication of relative performance.
## Load testing ## Load testing