OpsCenter performance metrics

In the Performance area of OpsCenter, you monitor a number of performance metrics about a Cassandra cluster. Real-time and historical performance metrics are available at different granularities: cluster-wide, per node, or per column family. OpsCenter 3.0 and later has been optimized to handle thousands of column families efficiently.

This section contains the following topics:

• Using performance metrics
• Cluster performance metrics
  • Write Requests
  • Write Request Latency
  • Read Requests
  • Read Request Latency
  • Cassandra JVM Memory Usage
  • JVM CMS Collection Count
  • JVM CMS Collection Time
  • JVM ParNew Collection Count
  • JVM ParNew Collection Time
  • Data Size
  • Total Bytes Compacted
  • Total Compactions
• Pending task metrics
  • Pending Task Metrics for Writes
  • Pending Task Metrics for Reads
  • Pending Task Metrics for Cluster Operations
• Column family performance metrics
  • CF: Local Writes
  • CF: Local Write Latency
  • CF: Local Reads
  • CF: Local Read Latency
  • CF: KeyCache Requests
  • CF: KeyCache Hits
  • CF: KeyCache Hit Rate
  • CF: RowCache Requests
  • CF: RowCache Hits
  • CF: Row Cache Hit Rate
  • CF: SSTable Size
  • CF: SSTable Count
  • CF: Pending Reads/Writes
  • CF: Bloom Filter Space Used
  • CF: Bloom Filter False Positives
  • CF: Bloom Filter False Positive Ratio
• System performance metrics
  • Memory
  • CPU
  • Load
  • Disk Usage (GB)
  • Disk Usage (%)
  • Disk Throughput
  • Disk Rates
  • Disk Latency
  • Disk Request Size
  • Disk Queue Size
  • Disk Utilization
  • Network Traffic