Title: The survey and future evolution of failure transparency and recovery techniques in Distributed Systems

Failure transparency (FT) meanings that system is available to user when a problem is occurring, the system still working without the user realizing it. FT is a measure that marks the availability of the system that normally works or works in a state of exception. Due to the emergence of different operating systems and Different platforms and need cover and fix errors in all forms in order to increase the reliability of systems absence a big blow to corporates credit, need to do a review of the latest scientific achievements in this area has been accessed. With the ever-increasing size of software and hardware components and the complexity of configurations, large-scale analytics systems face the challenge of frequent transient faults and permanent failures. As an indispensable part of big data analytics, MapReduce is equipped with a speculation mechanism to cope with run-time stragglers and failures. So in this article we have a solid understanding of the types of faults, and faults that led to failure, Techniques to prevent and recovery from failure.

کلیدواژه ها:

Software Failure ، Failure Transparency ، MapReduce ، Recovery for failure ، root cause of failure