Association for Computing Machinery
In this paper, the authors present BRRL, a library for making distributed main-memory applications fault tolerant. BRRL is optimized for cloud applications with frequent points of consistency that use data-parallelism to avoid complex concurrency control mechanisms. BRRL differs from existing recovery libraries by providing a simple table abstraction and using schema information to optimize checkpointing. They will demonstrate the utility of BRRL using a distributed transaction processing system and a platform for scientific behavioral simulations.