Efficient, balanced data placement algorithm in scalable storage clusters

Data distribution and load balancing become increasingly important in large-scale distributed storage system. This paper focuses on the problem of designing an optimal, self-adaptive strategies for balanced distribution and reorganization of replicated objects among a dynamically heterogeneous nodes, and presents a novel decentralized algorithm, Dynamic Interval Mapping, which maps replicated objects to a scalable collection of nodes, it distributes objects to nodes optimally, redistributing minimum amount of objects when new nodes are added or existing nodes are removed to maintain the balanced distribution. It supports weighted allocation and guarantees that replicas of a particular object are not placed on the same node. The time complexity and storage requirements are superior to previous methods.