Cluster Sizing

A full Welkin deployment requires a cluster with at least 40 CPUs and 82 GB of memory in total.

Monitoring

Monitoring stack (Thanos) can handle 6000 metrics per second in our standard configuration. This can be increased on demand.

Logging

Logging stack (OpenSearch) can take 100 records per second while provisioned with 12 CPUs and 24 GB of memory.