Cache and consistency in NOSQL
Cache and consistency in NOSQL
复制标题
NOSQL 中的缓存和一致性
DOI：
--
复制
发表时间：
2010
期刊：
影响因子：
--
通讯作者：
Zhi
中科院分区：
文献类型：
--
作者：
Peng Xiang;Ruichun Hou;Zhi
近年来，在持久性领域出现了一种新型的数据库（NOSQL）。2009年6月，一场全球性的NOSQL运动的聚会引发了“数据库革命”的导火索，非关系数据库现已成为一个极为热门的新领域，NOSQL旨在解决数据库的高并发读写、高效海量数据存储和访问、可扩展性和高可用性等需求。在这些大规模并发系统中，集群缓存和数据一致性成为关注的焦点。本文提出了一种快速有效的数据一致性估计方法。我们使用节点分裂和一致性哈希技术，以减少服务器的负载，通过分析来自客户端的请求。每个服务器节点可以被划分为一组虚拟节点，这些虚拟节点位于一个环中。集合的大小所引起的复杂性由目标系统决定。通过物理节点拆分，可以降低该高速缓存的未命中率。为了降低一致性哈希算法的复杂度，我们将介绍一个足球比赛的例子。
In recent years a new type of database (NOSQL) has emerged in the field of persistence. June 2009, a global gathering of NOSQL Movement triggered the fuse of “database revolution”, Non-relational database has now become an extremely popular new areas, NOSQL aims to solve the needs of high concurrent read-write, efficient mass data storage and access, database scalability and high availability. In these large-scale concurrent systems, cluster cache and data consistency become the focus of attention. This article proposes a fast and effective approach to estimate the solution of the data consistency. We use node split and consistent hashing techniques to reduce the servers load by analyzing the request from the client. Each server node can be divided into a set of virtual nodes, which located in a ring. The complexity caused by the size of the set is determined by the target system. Through physical node split, we can reduce the cache miss rate. In order to reduce the complexity of the algorithm of consistency hashing, we will introduce a football game example.