基于Redis实现HBase二级索引的方法

摘要

随着大数据时代的到来，海量数据对传统数据库技术提出了存储和检索性能的挑战。HBase是开源NoSQL数据库，适合于各种非结构化和半结构化的松散数据的存储和管理，目前已经被很多大型企业用于处理海量数据。它基于rowkey的有序存储，对rowkey支持毫秒级的快速检索。然而，随着HBase应用的不断深入，单一的通过rowkey检索数据的方式不再满足需求，在实际应用中，经常需要根据指定字段，或者几个字段进行组合检索。针对该问题，本文提出了一种基于Redis创建HBase 二级索引的方法，使得在实际应用中，支持多条件查询，提升查询的效率和性能。%With the arrival of the big data era, these massive data challenges traditional storage and retrieval per-formance for traditional database technologies. HBase is open source NoSQL database, suitable for a variety of un-structured and semi-structured data storage and management of the loose, and has been used by many large enterprises to deal with massive data. It is based on the rowkey of the orderly storage, supporting millisecond-level rapid retrieval based on rowkey. However, with the deepening of HBase applications, a single rowkey retrieval of data by no longer meet the needs of more applications. The practical applications often need to specify the field, or a combination of several fields to retrieve. To solve this problem, this paper proposes a method of creating HBase secondary index based on Re-dis, which can support multi-conditional query and improve query efficiency and performance in practical application.