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
You can either run all frameworks (takes about 16 hours) or just run selected frameworks: open a new console window cd webdriver-ts and keep that console window open
run e.g. `npm run bench keyed/vanillajs keyed/svelte keyed/solid keyed/react'
Create or copy an implementation and start developing in frameworks/keyed/${your-implemenation-name} (or non-keyed)
Open a separate console window, cd to that directory and keep it open
Use npm install (and delete node_modules and package-lock.json when appropriate to check that dependencies are really clean)
Make changes and recompile e.g. with npm run build-prod in that directory
Run a single benchmark (need to be run in webdriver-ts directory, use the console window you opened in "run peer group"):npm run bench -- --benchmark 03_ --framework keyed/${your-implemenation-name}. The benchmark runner logs the duration for each benchmark run, the mean, the median and so an. This might already be enough to see if the change helped. All benchmarks can be run with that number 01_, 02_, .. for the CPU benchmarks, 21_,22_ for the the mem benchmarks and 30_ for the startup benchmark (I know the number scheme isn't really logical).
Alternatively you can run the whole benchmark just for your framework npm run bench keyed/${your-implemenation-name} and update the table (run npm run results in webdriver-ts). Then you can open the table again.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
This discussion should help framework developers to find a good workflow to check the impact of changes:
Prepare:
npm ci && npm run install-local
unzip build.zip
npm start
and keep it running in the backgroundRun peer group:
cd webdriver-ts
and keep that console window opennpm run results
and open the results http://localhost:8080/webdriver-ts-results/table.htmlDevelop:
npm run build-prod
in that directorynpm run bench -- --benchmark 03_ --framework keyed/${your-implemenation-name}
. The benchmark runner logs the duration for each benchmark run, the mean, the median and so an. This might already be enough to see if the change helped. All benchmarks can be run with that number 01_, 02_, .. for the CPU benchmarks, 21_,22_ for the the mem benchmarks and 30_ for the startup benchmark (I know the number scheme isn't really logical).Alternatively you can run the whole benchmark just for your framework
npm run bench keyed/${your-implemenation-name}
and update the table (runnpm run results
in webdriver-ts). Then you can open the table again.Beta Was this translation helpful? Give feedback.
All reactions