[instrumentation-runtime-node] Add metrics from v8.getHeapStatistics

[tpilewicz]

Is your feature request related to a problem? Please describe

When using instrumentation-runtime-node, I want to be able to divide the metrics on heap usage by the heap size limit. This would allow alerting on the fact that e.g 80% of the heap size limit is used.

Describe the solution you'd like to see

If metrics were collected from v8.getHeapStatistics, I could divide for example used_heap_size by heap_size_limit to get heap utilization ratio.

Describe alternatives you've considered

I tried using the heap space statistics that the instrumentation records today, but they do not report on the actual hard limits there are on memory. For example this metric represents the space's current size, but the space might grow. So dividing heap.used by heap.limit doesn't achieve my goal.

Additional context

Happy to open a PR if it is deemed useful!

_{Tip: React with 👍 to help prioritize this issue. Please use comments to provide useful context, avoiding +1 or me too, to help us triage it. Learn more here.}

[danyalahmed1995]

I checked current main, and #3405 covers the heap_size_limit side by emitting v8js.memory.heap.limit from v8.getHeapStatistics().heap_size_limit.

One thing I’m not sure is fully covered yet: the issue mentions used_heap_size / heap_size_limit. The current v8js.memory.heap.used appears to be per-space rather than the global v8.getHeapStatistics().used_heap_size.

Aggregating the per-space v8js.memory.heap.used values considered sufficient here compared to global used-heap metric.

[tpilewicz]

Indeed the new meaning of v8js.memory.heap.limit matches my request. I also confirm that aggregating the per-space v8js.memory.heap.used tackles the use case of computing used / limit.

Closing the ticket then!