Deployment Topologies
TIBCO Fulfillment Order Management server side components (that is omsServer (including Orchestrator), AOPD, JeOMS, and OPE) and the client side user interface components (that is Configurator and omsUI with dashboard) must be deployed for leveraging their functionalities. Any one of the three topologies explained below can be followed for the deployment.
Single Node Single Instance Topology
In Single Node Single Instance topology, a single instance of one or more than one services run on the default ports.
This is the default topology that is supported immediately after the installation and initial configurations of Fulfillment Order Management.
Single Node Multi-Instance (Vertical Scaling) Topology
In the Single Node Multi Instance topology, the service instance under roles can be copied to create and start multiple instances. You may opt to create copies of all or a combination of the services and start them after changing their default port values. There should be exactly one instance of the Configurator service running in any kind of topology.
This topology is also referred to as Vertical Scaling. It allows the components to leverage the processing power efficiency of the machine. One micro-service instance runs in a single Java Virtual Machine process. However, the inherent concurrency limitations of a JVM process means that the components cannot fully utilize the processing power of the machine.
By running additional micro-service instances, multiple JVM processes are started and provide multiple thread pools. This allows the server components to utilize the maximum processing power of the machine.
Multi Node Multi-Instance (Horizontal Scaling) Topology
In Multi Node Multi-instance topology, multiple micro-service instances are started on multiple nodes to have multiple instances running in the cluster. You can deploy and start the application services on multiple nodes. The Configurator service must be installed and run only on one node. The other services can have multiple instances on any number of nodes.
This topology is also referred to as Horizontal Scaling. It allows the components to leverage the processing power efficiency of multiple machines. By running additional micro-service instances on multiple machines, multiple JVM processes are started and provides multiple thread pools. This allows the server components to use the maximum processing power of all the machines.