摘要
TPC-E是TPC(transaction processing performance council,事务处理性能委员会)组织发布的新一代数据库性能评测标准,用于取代日渐老化的TPC-C标准。SQLite是目前被广泛使用的一种嵌入式数据库,研究在TPC-E标准下的SQLite具体性能表现对实际应用具有重要参考意义。然而,目前业界缺少面向SQLite的TPC-E实现。为了解决该问题,设计和实现了面向嵌入式轻量级数据库SQLite的TPC-E基准测试系统,它不仅可以为SQLite数据库提供准确高效的性能评测系统,而且对SQLite在市场上的扩展应用和开发具有重要的参考价值。该测试系统严格遵循TPC组织最新发布的OLTP性能评测基准TPC-E,不仅能对SQLite作为内存数据库与文件数据库时的数据库特性进行精确有效的性能评测,还能客观真实地获得SQLite数据库的性能评测数据。实验结果表明,在TPC-E负载下,与Oracle等重量级的关系型数据库相比,SQLite事务处理性能较低。但由于其简单易用和体积小等优势,SQLite在实时性高的行业小型智能设备方面的应用依然具有较大的优势。同时通过对实验结果的分析,指出了制约SQLite性能的主要原因。
TPC-E,which is published by TPC （transaction processing performance council）,is the new generation of database performance evaluation standard to replace the aging TPC-C standard.SQLite is currently widely used as an embedded database in a lot of practical applications.Consequently,research on the specific performance of SQLite under TPC-E is of great reference to practical applications.However,the community lacks the TPC-E standard implementation for SQLite currently.To solve this problem,we design and implement a TPC-E benchmark for the embedded lightweight database—SQLite.It can not only provide an accurate and efficient performance evaluation,but also is of great reference value to the extended application and development of the SQLite database in the commercial market.This benchmark system strictly follows the TPC-E standard which is the newest testing benchmark of OLTP.It can evaluate the performance of the database features of SQLite as an in-memory database and a file database accurately and effectively,and can get the evaluation result in an objective way.Experiment shows that in the TPC-E standard the transaction processing performance of SQLite pales in comparison with heavyweight relational database such as Oracle.However,SQLite still has a greater advantage in the real-time industry,such as small-scale application in small intelligent devices due to their features of easy-using and small size.Besides,through the analysis of the experiment,we point out the main reasons that restrict the performance of SQLite.
作者
王龙翔
魏晓林
陈跃辉
刘成
董小社
WANG Long-xiang;WEI Xiao-lin;CHEN Yue-hui;LIU Cheng;DONG Xiao-she(School of Electronics and Information Engineering,Xi＇ an Jiaotong University,Xi＇ an 710049,China)
出处
《计算机技术与发展》
2018年第8期38-42,共5页
Computer Technology and Development
基金
国家重点研发计划(2016YFB1000303)
