You can optimize the performance of Pega Platform™ by using node classification, the process of separating nodes by purpose, predefining their behavior by assigning node types. With defined node types, your Pega applications can perform with maximum more efficiency; during the onboarding process, Pega helps you to determine what node types you need. The method of node assignment depends on your deployment:
- For cloud environments, Pega Cloud Services assigns nodes with node types during provisioning.
- For on-premises environments, which can mean either client-manage clouds or bare metal systems, you and your support team determine which node types you need. You assign the node types to nodes when you start the nodes.
For more general information, see the help overview, Classifying nodes.
Node behavior templates
With node classification, the type of node determines the work which that particular node type assignment performs in your Pega applications. Every node that is started with the same node type (using the same template) distributes the work loads over all of the nodes with the same node type assignment. While the node types are predefined for Pega software running in a Pega Cloud environment, in on-premises environments, you can assign a node with a node type during startup, which runs only those work work resources that are associated with that node type. You can associate the following work resources with one or more node types:
- Job Schedulers
- Queue Processors
By assigning node types to nodes and subsequently associating work resources with those node types, you can optimize resources and improve performance. Job Schedulers and Queue Processors require an association with a node type when you create these resources. For more information, see Creating a Job Scheduler rule and Creating a Queue Processor rule. To associate agents or listeners with a node type, see the corresponding pages: Associating agents with node types and Associating listeners with node types. To ensure your listeners are appropriately configured, see Best practices for processing files using the file listener.
Configuring node types
For on-premises environments, you configure a node with a node type during startup if you want to use node classification and run work resources according to their associations with a node type. If a node is not started with a node type, the system assigns one of the following node types: