Is your feature request related to a problem? Please describe.
Node.js (via V8) makes certain garbage collection metrics available to end users via a C++ based API. (see https://www.npmjs.com/package/gc-profiler). Having these metrics available to end users would be useful for users attempting to diagnose garbage collection related performance issues and failures in their applications.
Describe the solution you'd like
Open questions from product
What if services is having problems which could be identified using runtime metrics like eventloop, memory, GC, etc across all instances.
What if service is having those problems on few instances out of many?
When looking at a single instance - how to understand what the cause is? Or at least point in a direction
Is your feature request related to a problem? Please describe.
Node.js (via V8) makes certain garbage collection metrics available to end users via a C++ based API. (see https://www.npmjs.com/package/gc-profiler). Having these metrics available to end users would be useful for users attempting to diagnose garbage collection related performance issues and failures in their applications.
Describe the solution you'd like
Open questions from product
What if services is having problems which could be identified using runtime metrics like eventloop, memory, GC, etc across all instances.
What if service is having those problems on few instances out of many?
When looking at a single instance - how to understand what the cause is? Or at least point in a direction