A Failure Recovery Framework for Job Management Services

To solve the problem of service failure for job management services, a new failure recovery Framework named J-FRF is given. During the life cycle of service and job, states are monitored and contexts are saved. When service failure occurs, the framework takes advantage of the persistent contexts to realize rapid recovery of service and job. The experiments show that J-FRF has a high recovery success rate and lower recovery time, and can be able to better handle the failure of job management services.