随着Web2.0技术的高速发展,云计算中的大规模分布式服务和数据存储技术对传统的关系型数据库带来了巨大的挑战.NoSQL数据库打破了关系型数据的束缚,正在成为人们关注的焦点.NoSQL是一种非关系型数据库管理系统,松散的数据存储机制,不支持多表查询,有高效的查询功能.文中首先介绍了MongoDB数据库自动分片的原理和实现机制,然后为了解决在自动分片中数据负载不均衡,提出了基于数据操作频率的改进算法.这个改进的平衡策略可以有效地均衡分片中的数据,提高集群的并发读写性能.%With the rapid development of the Internct Web2.0 technology,the demands of large-scale distributed service and storage in cloud computing have brought great challenges to shackles of RDBMS is becoming the focus of attention.NoSQL is non-relational database management system with format loose data storage,not supporting the join operation,the effective query capability etc.In this paper,the principles and implementation mechanisms of auto-sharding in MongoDB databases are firstly presented,then an improved algorithm based on the frequency of data operation is proposed in order to solve the problem of uneven distribution of data in auto-sharding.The improved balancing strategy can efficiently balance the data among shards,and improve the cluster's concurrent reading and writing performance.
展开▼
