Resource Requirements for Self-Managed Connectors

To achieve the desired performance and throughput, your hosting infrastructure must provide sufficient resources to your Self-Managed Connectors. The resource requirements vary and depend to a large extent on the number of workflows you are running and the batch size you have configured. The minimum requirements are summarized below.

Java Development Kit (JDK)

The Self-Managed Connectors require a minimum of JDK 17.

Host Infrastructure

We recommend that you provision a dedicated virtual machine or compute instance for each connector instance.

For Kubernetes, we recommend that you deploy each connector container instance in its own pod.

The host environment should allocate the following resources to the connector instance:

• Multi-core CPU. Each workflow uses a minimum of two threads. We recommend allocating as many cores/vcores as possible, especially if you are configuring more than one workflow per connector instance.

• 2.0 GB RAM

• 5.0 GB minimum disk space. Extra disk space may be required depending on your logging configuration.