graph-node
graph-node copied to clipboard
chore(ci): query performance testing using K6
- [x] Setup a new entry point for running
graph-nodeand all requirements - [x] Reuse parts from e2e/integration testing (moved shared parts into a new
crate) - [x] Setup a test project with a contract + Subgraph
- [x] Run K6 as part of the CI workflow
- [x] Add GH Actions workflow for running + commenting results on each PR.
✅ Benchmark Results
✓ no_errors
✓ expected_result
checks.........................: 100.00% ✓ 18210 ✗ 0
data_received..................: 5.3 MB 177 kB/s
data_sent......................: 2.8 MB 93 kB/s
http_req_blocked...............: avg=4.44µs min=800ns med=1.8µs max=1.78ms p(90)=2.8µs p(95)=3.4µs
http_req_connecting............: avg=1.01µs min=0s med=0s max=1.54ms p(90)=0s p(95)=0s
http_req_duration..............: avg=32.69ms min=8.01ms med=31.2ms max=101.97ms p(90)=44.42ms p(95)=49.82ms
{ expected_response:true }...: avg=32.69ms min=8.01ms med=31.2ms max=101.97ms p(90)=44.42ms p(95)=49.82ms
http_req_failed................: 0.00% ✓ 0 ✗ 9105
http_req_receiving.............: avg=57.63µs min=11.5µs med=31.7µs max=10.19ms p(90)=82.16µs p(95)=140.98µs
http_req_sending...............: avg=25.23µs min=5.5µs med=10.1µs max=6.72ms p(90)=18µs p(95)=36.37µs
http_req_tls_handshaking.......: avg=0s min=0s med=0s max=0s p(90)=0s p(95)=0s
http_req_waiting...............: avg=32.61ms min=7.97ms med=31.11ms max=101.91ms p(90)=44.34ms p(95)=49.63ms
http_reqs......................: 9105 303.315348/s
iteration_duration.............: avg=32.94ms min=8.11ms med=31.43ms max=102.14ms p(90)=44.77ms p(95)=50.17ms
iterations.....................: 9105 303.315348/s
vus............................: 10 min=10 max=10
vus_max........................: 10 min=10 max=10
@dotansimha, is this still WIP? I mean by the commit title and message.
@dotansimha, is this still WIP? I mean by the commit title and message.
No longer WIP, I rebase and resolved your review comments, thanks!
@dizzyd I saw you self-assigned. I can rebase this PR if you want to get this in soon :)
Hey @dotansimha - I'm going to close this PR, as we talked about breaking it up into some smaller PRs. Let me know if you want to re-open/change this one -- that's fine too. :)