I am ovarall happy with this PR except for naming. We need to get the naming right now

I applied one of the suggested names. Let me know if folks prefer something else.

@bbakerman, @andimarek: I have updated the PR to make the new instrumentation specific, as well as annotating it with @ExperimentalApi now to avoid consumers taking a hard dependency on it.

@timward60 just to make you aware: we are adding a profiler: #3976 This will be merged very soon. Maybe if the only goal of the instrumentation is to measure peformance we should consider using Profiler. What do you think?

@timward60 just to make you aware: we are adding a profiler: #3976 This will be merged very soon. Maybe if the only goal of the instrumentation is to measure peformance we should consider using Profiler. What do you think?

I see, let me take a look at that proposal/implementation and get back to you. In general we use the instrumentation callbacks for profiling in many places e.g. logs, open telemetry tracing, metrics (we have various internal systems for different diagnostic aspects). I would need to see if it would still align and that we could still expose it those various consumers.

If there is some reluctance in the meantime, I will hold off on this PR until I have time to get back to you on if I foresee any issues with the profiler proposal from our usage above. In the meantime we will continue to use our internal fork to collect some data. Main reason behind this is to measure the move from ExecutableNormalizedOperations to NormalizedDocuments (with caching support). Once I add some test coverage to the draft PR (#4048), I will publish it for review.

Hello, this pull request has been inactive for 60 days, so we're marking it as stale. If you would like to continue working on this pull request, please make an update within the next 30 days, or we'll close the pull request.

Commenting to keep open

Hi @timward60, would you like to continue working on this PR? Totally ok if you are busy, wanted to check and see how you're going