Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
For getting stable results we need to run performance tests many times and then calculate mean, median and standard deviation for results. luatest has an option `-r N` that allows to run testcase N times, but it don't execute setup/teardown on each iteration [1] and due to this our testcases fails due to duplicated id's in a space. For running tests many times I made an external loop in a shell script: ``` for i in `seq 1 1 30`; do luatest -v -c test/performance/select_perf_test.lua 2>&1 | grep -E 'INSERT|SELECT'`; done ``` For more convenient extracting numbers from log I changed format of printing on each iteration - all numbers that changed from run to run printed in a single row, and removed printing numbers that are always static. 1. tarantool/luatest#161
- Loading branch information