Adjust the scale factor to modify the number of worker threads per app node, which increases or decreases throughput.

Before you begin

Confirm the following:
  • Discovery Admin Workspace is using version 1.9.0 or later.
  • ServiceNow AI Platform® is using version XP9, YP4, or later than YP4.
  • The discovery.use.event.processing system property is turned on. For more information, see Configure Discovery to use Event Framework.
  • The Job configuration type field of the Queue Registration form is set to Scale with nodes. For more information, see Queue Registration form reference.

Role required: discovery_admin

About this task

If your instance becomes slow while Discovery is running, you can adjust the scale factor to modify the number of worker threads per app node. The default scale factor is set to 2, which corresponds to two worker threads per app node.

For example, if you increase the scale factor to 3, Discovery uses three worker threads per app node, increasing throughput. Conversely, reducing the scale factor to 1 limits Discovery to a single worker thread per app node, which decreases throughput.

Procedure

  1. Navigate to All > System Policy > Queue Registry.
  2. Choose one or more job types for which you want to adjust the scale factor.
    Job typeAction
    IP-based Select discovery.sensors from the Queue Registry [queue_registry] table.
    Cloud-based Select discovery.cloud.sensors from the Queue Registry [queue_registry] table.
  3. In the Scale factor field, set a value.
    Note: The default maximum value is 3. To increase the scale factor beyond 3, configure the com.glide.processing.framework.max_thread_utilization_percentage property. For more information about this property, see Discovery properties. For instructions on how to create a property, see Add a system property.
  4. Select Update.