Collecting object I/O level diagnostics

Steps to collect data on object I/O statistics using the DataStax Enterprise Performance Service.

The following object I/O level diagnostic tables collect data on object I/O statistics:
  • object_io

    Per node recent latency metrics by keyspace and table.

  • object_read_io_snapshot

    Per node recent latency metrics, broken down by keyspace and table and orders data by mean read latency.

  • object_write_io_snapshot

    Per node recent latency metrics, broken down by keyspace and table and orders data mean write latency.

Procedure

To enable the collection of this data:

  1. Edit the dse.yaml file.

    dse.yaml

    The location of the dse.yaml file depends on the type of installation:

    Package installations
    Installer-Services installations

    		 /etc/dse/dse.yaml

    Tarball installations
    Installer-No Services installations

    		 installation_location/resources/dse/conf/dse.yaml
  2. In the dse.yaml file, set the enabled option for resource_level_latency_tracking_options to true. 
    # Data Resource latency tracking settings
resource_level_latency_tracking_options:
    enabled: true
    refresh_rate_ms: 10000
  3. (Optional) To control how often the statistics are refreshed, increase or decrease the refresh_rate_ms parameter.

    The refresh_rate_ms specifies the length of the sampling period, that is, the frequency with which this data is updated.