Instrument disk throughput, await and utilization metrics - Githubissues

opensearch-project / OpenSearch

🔎 Open source distributed and RESTful search engine.

https://opensearch.org/docs/latest/opensearch/index/

Apache License 2.0

9.77k stars 1.82k forks source link

Instrument disk throughput, await and utilization metrics #1125

Open sachinpkale opened 3 years ago

sachinpkale commented 3 years ago

Is your feature request related to a problem? Please describe.

Existing fs stats provides I/O related metrics from /proc/diskstats file but these metrics provide current value of the metrics (like read_operations, write_operations etc) and it is difficult to understand a potential issue just by looking at these metrics.
Client needs to maintain a state to understand if current value is less/more than previous value by a certain threshold.

Describe the solution you'd like

Opensearch maintains a state with previous values and create metrics that can be used directly.
Few metrics that can be derived from the values of /proc/diskstats are:
- Disk throughput
- Disk Await
- Disk utilization
Throughput and utilization metrics would help to find potential issues and mitigate them faster.

anasalkouz commented 2 years ago

I would prefer to keep such metrics out of the core engine, it might be a good idea to extend the functionality using a plugin. Move this to pending more input from community in order to prioritized it.