Profiling lage
A particularly complex monorepo can present opportunities for optimization. For example, when there are really large packages, it might be more efficient to break those up so the build can be split across different CPU cores. lage greatly enhances the ability for developers to see where the bottlenecks are.
To collect a profile of the lage run, simply add the --profile argument.
Using the profile
When the run is finished, a profile JSON file is produced. This file is to be imported into a Chromium-based browser's devtools Performance tab.
Sample of lage profile session
For example, you can see the following lage run of the Fluent UI repo. You can see that there is a dip in the concurrency when building the office-ui-fabric-react package. This makes sense, because many internal packages depends on that one large package.