Scalable and Decoupled Logging for State Machine Replication

State Machine Replication (SMR) is a widely used approach for fault tolerance of important services. Support for SMR implementations on shared infrastructures has emerged, allowing wider adoption. However, there are still non-trivial aspects that developers have to handle to build and deploy their dependable services. In this paper, we tackle the need for recovery to keep fault-tolerance levels, and propose an approach to: (i) simplify the development of logging; (ii) improve resource sharing in shared infrastructures; (iii) alleviate costs with replication in pay-per-use infrastructures. The central idea is to decouple service execution from logging and offer logging functionality as a service attachable to SMR deployments. Beyond the added simplicity to deploy an SMR, we show that this approach does not penalize performance of replicated services, and that a logging service can scale to look to several applications.

Local and Remote Recovery of Cloud Services Using Backward Atomic Backup Recovery Technique for High Availability in Strongly Consistent Cloud Service

Data loss occurs due to crashing, correlated failure, logical failure, power outages and security threats. Several techniques (e.g. NoBackup, WARBackup and LocalRecovery) are being used to recover data locally. And, strongly consistent Cloud services (SCCS) must provide good performance and high availability. However, conventional strong consistency replication methods have the limitation of availability of replicated services when recovering huge amount of data across wide area links. There is a need for remote recovery mechanisms for high availability of service/data, because distributed nature of cloud infrastructures. To address these issues, the article proposes a hierarchical system architecture for replication across a data center, and employs the backward atomic backup recovery technique (BABRT) for local recovery and remote recovery for high availability of the cloud services/data. A mathematical model for BABRT is described. Simulation results show that BABRT reduces the storage consumption, recovery time, window of vulnerability and failure rates, compared to other recovery models.