---
sourceDocument: Australia ServiceNow AI Platform Capabilities
sourceDocumentLink: https://servicenow-prod.fluidtopics.net/r/de-DE/servicenow-platform

 Release :

    - australia

ft:locale :

    - de-DE

ft:publication_title :

    - Australia ServiceNow AI Platform Capabilities

ft:clusterId :

    - platcap

bundleId :

    - platcap

workflow :

    - Platform


---

# Partition size computation

# Partition size computation for parallel loading in Integration Commons for CMDB {#ariaid-title1}

* Freigeben Version: Australia
* 
* Aktualisiert 12. März 2026
* 
* ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 Minute Lesedauer

Integration Commons for CMDB supports computed partition size for the parallel loading framework, if supported in a Service Graph Connector.

The parallel loading feature of ServiceNow AI Platform, executes data load using multiple threads. The partition size for a data source is computed by the Integration Commons for CMDB application used by the parallel loading feature. This partition size confirms that the workload is evenly distributed among the available threads. By balancing the data load, a data source
within a Service Graph Connector can manage data loading uniformly across the available nodes, reducing processing time and improving overall efficiency. This approach promotes efficient data loading, minimizes latency, and optimizes the
performance of data sources.  
The partition size computation for the parallel loading feature within the Integration Commons for CMDB application is based on the following calculations:

* [Partition size calculation](https://servicenow-prod.fluidtopics.net/gpNK8109ha75KnaPMijChw#integration-commons-part-size__section_gst_h2g_ccc)
* [Parallel jobs calculation](https://servicenow-prod.fluidtopics.net/gpNK8109ha75KnaPMijChw#integration-commons-part-size__section_zfw_z2g_ccc)
{#integration-commons-part-size__ul_uv4_b3g_ccc}

## Partition size calculation {#integration-commons-part-size__section_gst_h2g_ccc}

The partition size for parallel loading in a data source is determined by both the number of active primary nodes on your ServiceNow instance and the total volume of data to be loaded by the data source using the following formula:

    Partition size = Total data volume / (10% * Active primary nodes)

For example, if there are `4` active primary nodes and the data volume to be loaded is `1000` units, the partition size is calculated as:

    Partition size = 1000 / (10% * 4) = 1000 / 0.4 = 25

## Parallel jobs calculation {#integration-commons-part-size__section_zfw_z2g_ccc}

Parallel jobs enable distributing the workload evenly across multiple jobs, reducing the data loading time and improving the efficiency of data retrieval. The calculation of the number of parallel jobs created for a partition is based on the following formula:

    Number of parallel jobs = Total data volume/Partition size

In the earlier example, where the total data volume is `1000` units and the partition size is `25`, the number of parallel jobs created is calculated as:

    Number of parallel jobs = 1000/25 = 40