Set partition size for parallel loading-enabled AWS accounts

  • Release version: Australia
  • Updated March 12, 2026
  • 1 minute to read

    • When using the SG-AWS-Image-Private, SG-AWS-Tags, SG-AWS-SendCommand, and AWS-SSM-GetS3Object data sources, you can optimize their performance by setting the partition size in the aws_account_partition_size connection property.

    Before you begin

    Set the application scope to Service Graph Connector for AWS using the application picker. For more information, see Application picker.

    Role required: admin

    About this task

    The parallel loading feature is available for the SG-AWS-Image-Private, SG-AWS-Tags, SG-AWS-SendCommand, and AWS-SSM-GetS3Object data sources from the Washington DC Patch 5 release. This feature determines the number of AWS accounts that can be executed as parallel jobs to retrieve configuration item (CI) data. Partition size determines how many AWS accounts a single parallel-thread evaluates for data retrieval. For example, if you set the aws_account_partition_size property to 5 and there are 25 AWS accounts, the application generates 5 parallel jobs during data loading.
    Note:
    If the aws_account_partition_size property isn't configured, the partition size computation feature of the Integration Commons for CMDB application automatically determines the partition size for parallel loading at runtime. See Partition size computation for parallel loading in Integration Commons for CMDB.

    Procedure

    1. Navigate to All > Service Graph Connectors > AWS > Connections.
    2. From the Name column of the Service Graph Connections list, select a connection.
    3. From the Property column of the Service Graph Connection Properties related list, select aws_account_partition_size.
    4. In the Value field, enter a numeric value greater than 0 for the partition size.
    5. Select Update.