MapR Database Metrics

Every 10 seconds, the collectd service uses a MapR plugin to gather MapR Database metrics on each node in the cluster. MapR Database provides both node and table metrics.

Node metrics capture data for operations across a MapR node. You can use them to assess the performance of individual nodes in your MapR cluster.

Starting in MapR 6.1, MapR Database supports table metrics. Table metrics provide more granular metrics. They allow you detect and diagnose bottlenecks and performance issues that are specific to individual tables. For example, suppose a node metric shows a spike in a particular RPC. Knowing this, you can use the corresponding table metric to determine if the spike originates from a single table.

Examples of other use cases that benefit from table metrics are the following:

  • You want to measure the latency of different RPC operations on a table
  • You want to determine which operations on a table are slow
  • You want to determine which tables are most frequently accessed