一种基于数据热点的MongoDB负载均衡策略

MongoDB作为一种新兴的NoSQL数据库，以其模式自由、文档式存储、故障自动恢复、良好的水平扩展、自动负载均衡等特点深受国内外市场的青睐.MongoDB自带的负载均衡策略能使各个节点数据量达到平衡.但是在实际的生产环境中，节点之间数据访问热度不同也会导致负载失衡，特别是出现节点过热的情况.针对这一问题，引入Markov随机过程，提出一种基于Markov预测模型的负载均衡策略，根据Markov模型的稳态概率向量预测各个分片的负载并进行数据迁移.通过实验，验证了当各个分片间出现节点过热时，所提出的负载均衡策略能够很好地使分片间的负载达到基于访问热点的均衡.

Load Balancing Scheme for MongoDB Based on Data Hotspot

As one of the new NoSQL databases, MongoDB has attracted a great deal of attention in domestic and foreign market due to its prominent features such as schema-free, document-oriented, fault auto-recovery, horizontal scaling and auto-load balancing. The auto-load balancing strategy guarantees that the stored data can be distributed to all nodes evenly without taking the data hotspot into consideration. However, load imbalance can occur among the nodes based on hotspot in the actual production environment, which especially generates hot nodes. To address this problem, the Markov stochastic process is introduced by presenting the load balancing strategy based on the Markov model. The chunk loads are predicated by the vector of stable probability before data migrate across the chunks. The experimental results show that the migration strategy proposed in this paper can balance the clusters effectively based on data hotspot when hot nodes get overloaded.