Lesson 6 Overview

TIBCO SmartSockets is designed to detect and handle many different kinds of network failures, and this robustness provides a high-level of fault tolerance. Fault tolerance is a term used to describe computer systems that continue to function even when some of the hardware or software fails. Fault tolerance can be implemented in hardware by mirrored file systems on multiple disks, redundant networks, redundant CPUs, redundant memory, and so on. Hardware-based fault tolerant systems are more expensive than non-fault tolerant systems due to the extra components. Fault tolerance can also be implemented in software by products such as TIBCO SmartSockets. Surprisingly, enabling the fault tolerant features of TIBCO SmartSockets has little effect on message throughput. The general mechanisms that TIBCO SmartSockets uses for fault tolerance are:

avoiding operations that can block indefinitely, or putting an upper limit on the amount of time these operations can block

periodically checking for potential failure conditions using keep-alives

auto-starting and restarting of RTserver; for example, if RTserver fails while the system is running, RTclients can be set to automatically try to re-start RTserver

RTservers can reconnect to each other if network failures cause a temporary loss of connectivity

hot failover of RTservers

hot failover of RTclients

message buffering; for example, if a message arrives when an RTclient is not ready or is behind, the message is buffered until the process is ready

guaranteed message delivery (GMD), messages can be delivered even in the face of network and system failures

dynamic message routing where messages can take alternative routes depending on what parts of the network is up

real-time monitoring that provides instant notification of important events

This lesson covers a few of the more commonly used methods to build a robust and reliable distributed application in TIBCO SmartSockets.

Lesson 6 Overview

Fault Tolerance