结合HBase的散列概要森林索引方案

Hash Synopsis Forest Index Schema Based on HBase

时序数据在各个领域的使用越来越广泛,例如在金融领域回撤操作中对时序数据的聚合操作也就是时序数据库的重要的应用场景.传统关系型数据库不足以支撑海量时序数据,而现有的No SQL数据库对时序数据的聚合操作显得低效耗时.概要森林是一种种结合概要表和线段树思想的支持时序数据聚合操作的高效索引机制.本文结合分布式No SQL数据库HBase的机制,实现一种基于HBase的散列概要森林的索引方案,优化聚合操作速度,并提出一种基于HBase的线段树散列存储方案,解决HBase分布式存储时序数据产生热点问题.最后通过与通用开源时序数据库Opentsdb以及一般索引机制查询对比实验,验证了该索引机制的可用性和高效性.

Time-series data is a key to every fields,especially aggregate operation of Time-series data in the financial field. However,traditional relational databases fail to support vast amounts of time-series data. The NoSQL database are inefficient for time-series dataaggregation. Synopsis Forest are an efficient indexing mechanism that supports the aggregation of time-series data by combining synop-sis tables and segment tree. This paper presents an efficient index mechanism that support time-series data aggregation base on synopsisforest which is a efficitive time-series data aggregation algorithm. Then combine HBase mechanism optimize the aggregation queryspeed ,design a hash mechanism to solve HBase distributed hot pot problem. Design experiment to compare this index mechanism withOpentsdb and original HBase aggregation query.