chmonitor vs Nova

chmonitor reads ClickHouse system tables to show queries, merges, replication and disk health in one place, with alerting and a read only AI advisor.

Nova is the analytics and orchestration layer for AI products. Track usage, evaluate quality, and iterate on prompts, all in one dashboard.

About chmonitor

chmonitor is an open source operations dashboard for teams running ClickHouse. It builds its views directly from ClickHouse system tables, so no exporters or extra storage are required. Engineers can inspect running, failed, slow and costly queries, run EXPLAIN, and stop long queries from the table. A topology map shows nodes, shards and replicas plus the Keeper quorum with metrics for each node, while health checks notify Slack, Discord, PagerDuty or Opsgenie when thresholds are crossed. A built in agent studies schema and query history to suggest tuning, and it can be reached over MCP in read only mode. Self hosting is free, and a CLI is in beta.

About Nova

Nova is the analytics and orchestration layer for AI products. Track usage, evaluate quality, and iterate on prompts, all in one dashboard. Nova gives you the observability layer your AI features deserve. Run controlled experiments across prompt variants and route traffic to the winner automatically. By the end of the day we had dashboards our whole team actually uses. You can redact PII at the SDK level before anything leaves your servers. Overage is billed at $1 per 10k events, and you can set a hard cap if you prefer strict budgets.