You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
|`TAG.spec`|`'spec-benchmark'`| Special tag that marks a benchmark as a spec-required benchmark |
73
+
|`TAG.alert`|`'alerting-benchmark'`| Special tag that enables our perf monitoring tooling to create alerts when regressions in this benchmark's performance are detected |
74
+
|`TAG.cursor`|`'cursor-benchmark'`| Tag marking a benchmark as being related to cursor performance |
75
+
|`TAG.read`|`'read-benchmark'`| Tag marking a benchmark as being related to read performance |
76
+
|`TAG.write`|`'write-benchmark'`| Tag marking a benchmark as being related to write performance |
77
+
61
78
## Wishlist
62
79
63
80
- Make it so runner can handle: `./lib/suites/multi_bench/grid_fs_upload.mjs` as an argument so shell path autocomplete makes it easier to pick a benchmark
exportconsttaskSize=3.1401000000000003;// ~3MB worth of work
12
-
13
-
assert.equal(taskSize,byteLength*10e-6);// taskSize should stay hardcoded, checking here the math is done right.
14
9
10
+
conststableRegionMean=42.82;
11
+
exportconsttaskSize=3.1401000000000003/stableRegionMean;// ~3MB worth of work scaled down by the mean of the current stable region in CI to bring this value to roughly 1
0 commit comments