The need for guaranteed QoS and efficient management in Service Oriented Infrastructures is an essential requirement for the deployment, execution, and management of modern business applications. In that frame, the capabilities for fault detection and recovery in all layers of a Service Oriented Infrastructure are essential for the smooth operation of the business applications and the wide adoption of these solutions in the global market. In this chapter, we present the concepts of fault detection and recovery, including terminology, classification of faults, and analysis of the key processes taking place in a system in order to diagnose and recover from failures. The state of the art mechanisms and techniques for fault detection and recovery are also analyzed, while recommendations for applying them in Service Oriented Infrastructure are presented.
TopIntroduction
Service Oriented Infrastructures, and generally the Distributed Systems, are increasingly considered to be ideal candidates for implementing platforms capable to support business applications. This is tightly coupled with the need for high quality of service (QoS) provisioning for the applications not only to guarantee their smooth operation and management but also to support the primary capabilities of such systems for scalability, service orchestration, autonomous management and abstraction of the systems’ complexity. In this frame, fault tolerance functionality is necessary to achieve high QoS provisioning for both the system and the applications. The advantage of Service Oriented Infrastructures from the fault tolerance perspective is significant. These systems can be easily made redundant, which is the corner-stone for all fault tolerance techniques. Unfortunately, distribution also means that the imperfect and fault prone physical world cannot be ignored, so that as much as they help in supporting fault tolerance, distributed systems may also be the source of many failures.
On the other hand, the building blocks and components of the SOIs are independent of each other and therefore are independent points of failure. Even though this may be an advantage for the end users, it is also a complex problem for the developers and administrators in terms of synchronization and management. In fault tolerant distributed systems, a component failure means that the other components and services have to detect and handle that failure to keep the system running and maintain the QoS level for the applications and end users to an acceptable level. This involves redistributing the functionality from the failed component to other, or it may mean switching to some king of emergency mode for the operation of the systems and the execution of the application.
Providing fault tolerant design for every component is normally not an option. In such cases the following criteria may be used to determine which components should be fault tolerant:
- •
How critical is the component? In a car, the radio is not critical, so this component has less need for fault tolerance.
- •
How likely is the component to fail? Some components, like the drive shaft in a car, are not likely to fail, so no fault tolerance is needed.
- •
How expensive is it to make the component fault tolerant? Requiring a redundant car engine, for example, would likely be too expensive both economically and in terms of weight and space, to be considered.
In the next sections of introduction we present the terminology related with fault detection and fault tolerance systems since many of these terms are confused. In addition we have categorized the faults that may occur in SOIs based on their duration, cause and effect and we analyze their most important drawbacks. In section two, the key processes that take place in such systems to provide the fault tolerance functionality are described. Section three presents the state of the art mechanisms and techniques for the fault tolerance while section four concludes our work and highlights the future trends.