Transactional computation on clusters

摘要: BACKGROUNDDistributed computing is the use of multiple autonomous computers (or processors) to solve computational problems by dividing the problem into many sub-problems that are then Solved by one or more of the autonomous computers (or nodes) in a cluster of computers. Distributed shared memory (DSM) is an aspect of distributed computing where each node of a cluster has access to shared memory that is distributed across the cluster and which may comprise local memory, remote memory, or both with regard to each node. Certain performance-critical workloads—such as those requiring more memory or processing power than is available in a single computer—can benefit from the abstraction of a mutable (changeable) shared State that is made possible via distributed computing systems employing distributed shared memory.To perform computations on very large datasets, distrib uted computing …