Monitoring overview
Imply Polaris includes built-in monitoring to support your production workloads and ensure your database is optimized for your application. The Monitoring section of the Polaris user interface provides dashboards for inspecting query performance and event stream ingestion. Use the provided information to perform comparative analyses for the entire cluster to see how performance has evolved over time. You can also probe into detailed metrics in the prepared data cube interface. Access these dashboards and visualizations in the Monitoring section of the left navigation tree.
This topic provides an overview of the monitoring capabilities included in Polaris.
User queries
The User Queries view provides a single-page dashboard for monitoring query performance. In this dashboard, you can analyze the following:
- Query performance: Track query latency, total number of queries executed, and total number of failed queries.
- Query processing: Evaluate the average and 98th percentile of query wait times to determine whether to scale up your project in response to high concurrent load issues.
- Segment scanning: Assess the number of segments scanned and the segment scan times. High segment scan times indicate that your segment files are too large, which can be resolved by data partitioning. If many scans are occurring, your data may be too fragmented, and you may benefit from configuring data rollup.
Streaming
The Streaming view provides a dashboard to monitor streaming ingestion. This dashboard displays the following:
- Volume of incoming events and latency to ingest those events
- Issues from streaming ingestion, including unparseable events and expired records rejected by Polaris
- Number of rows output from processed events
Detailed metrics
The Detailed Metrics view provides a data cube where you can investigate specific metrics with the option to filter by table, query type, and query ID.