Low cost management of replicated data in fault-tolerant distributed systems

TLDR

Distributed systems replicate data for fault tolerance and availability, but synchronizing updates across copies incurs communication delays that slow operations. The paper proposes a technique that relaxes synchronization to allow concurrent updates of replicated data while preserving correctness. The technique permits concurrent updates by relaxing synchronization constraints and is compatible with roll‑back and roll‑forward failure recovery mechanisms. The relaxed synchronization yields faster response times for operations on replicated data.

Abstract

Many distributed systems replicate data for fault tolerance or availability. In such systems, a logical update on a data item results in a physical update on a number of copies. The synchronization and communication required to keep the copies of replicated data consistent introduce a delay when operations are performed. In this paper, we describe a technique that relaxes the usual degree of synchronization, permitting replicated data items to be updated concurrently with other operations, while at the same time ensuring that correctness is not violated. The additional concurrency thus obtained results in better response time when performing operations on replicated data. We also discuss how this technique performs in conjunction with a roll-back and a roll-forward failure recovery mechanism.