Determine what kind of metrics should be emitted so that external resource schedulers (e.g., HPA, KEDA, etc.) can add or remove compute resources based on demand. For example, external compactor scaling based on compaction queues. Or, scaling the number of ScanServers based on current ScanServer load and number of clients performing scans.
Determine what kind of metrics should be emitted so that external resource schedulers (e.g., HPA, KEDA, etc.) can add or remove compute resources based on demand. For example, external compactor scaling based on compaction queues. Or, scaling the number of ScanServers based on current ScanServer load and number of clients performing scans.