A survey on checkpointing web services

Abstract

With the advent and widespread use of web services, complex business processes are being built by discovering and composing services already available over the Internet. Such composite web services operate over the Internet where reliability and speed cannot be guaranteed, hence providing fault handling mechanism to composite web services is of primary importance. Several fault handling and recovery techniques have been proposed in literature in the context of web services. Backward recovery using checkpointing is a general recovery scheme that can be used to make web services resilient to faults. Checkpointing strategies proposed in distributed systems are not directly applicable to web services due to the differences in the two paradigms. Few papers have been published discussing the need and techniques for checkpointing web services. In this paper we present a survey on various distributed and web service checkpointing techniques discussing their applicability, strengths and weaknesses. We give a brief introduction to our approach of checkpointing web services which identifies checkpointing locations, without user intervention, using complexity of interactions and services offered.