High Availability Deployment Of Runtime Components

Fault tolerance and high availability configuration setup of runtime components of API Exchange Gateway software.

This section describes the deployment of runtime components in a high availability setup. The runtime components are Core Engine, Cache Agents, Global Throttle Manager, Central Logger and Cache Cleanup Agents. You should deploy the runtime components in such a way that they are highly available in production systems to achieve maximum functionality of the gateway.

TIBCO API Exchange Gateway is deployed as a cluster of Core Engines that together act as a single logical gateway. The Core Engines in the cluster can run on a single server or in a distributed environment across multiple physical or virtual servers.

Typically for production deployment requirements, you must add additional instances of Core Engines. The architecture of Core Engine has been designed so that when multiple Core Engine instances are deployed in a gateway cluster, the key management functions such as global throttle management, cache management, cache cleanup management and central logging are coordinated across all Core Engine instances.

However, as the levels of transactions increase, it is likely that there is a corresponding increase in management activity. To avoid the possible impact of management activity upon the Core Engines of the TIBCO API Exchange Gateway, the management components (Global Throttle Manager, Cache CleanupAgent, and Central Logger) and the TIBCO Spotfire servers should be moved onto separate servers.

If multiple instances of the Core Engines are deployed in a cluster, you must start the Cache Agent instances explicitly once you start the Core Engines.

TIBCO Rendezvous is used for the communication between most of the runtime components of the gateway. The runtime components share a single set of configuration files.