华东师范大学学报（自然科学版） ›› 2023, Vol. 2023 ›› Issue (5): 1-10.doi: 10.3969/j.issn.1000-5641.2023.05.001

• 数据库系统 •    

基于持久化内存和共享缓存架构的高性能数据库

王聪聪, 胡卉芪*()   

  1. 华东师范大学 数据科学与工程学院, 上海　200062
  • 收稿日期:2023-07-25 接受日期:2023-07-25 出版日期:2023-09-25 发布日期:2023-09-15
  • 通讯作者: 胡卉芪 E-mail:hqhu@dase.ecnu.edu.cn
  • 基金资助:
    上海市自然科学基金(23ZR1418300)

Persistent memory- and shared cache architecture-based high-performance database

Congcong WANG, Huiqi HU*()   

  1. School of Data Science and Engineering, East China Normal University, Shanghai　200062, China
  • Received:2023-07-25 Accepted:2023-07-25 Online:2023-09-25 Published:2023-09-15
  • Contact: Huiqi HU E-mail:hqhu@dase.ecnu.edu.cn

RichHTML

23

PDF

99

摘要：

云原生数据库的崛起, 使得共享缓存架构再度受到重视. 虽然共享缓存架构能够有效地解决多个读写节点间的缓存一致性问题, 但其他问题仍然存在, 例如持久化速度缓慢、维护缓存目录的延迟高、时间戳瓶颈等. 针对上述问题, 提出了一种基于共享缓存架构, 并结合新型硬件—持久化内存的解决方案, 从而构建了一个包括内存层、持久化内存层、存储层的三层共享架构数据库—TampoDB. 基于此架构, 重新设计了事务的执行流程, 并对时间戳和目录进行了优化, 以解决上述问题. 实验结果表明, TampoDB有效地提高了事务的持久化速度.

关键词: 云原生数据库, 持久化内存, 共享缓存

Abstract:

The upsurge in cloud-native databases has been drawing attention to shared architectures. Although a shared cache architecture can effectively address cache consistency issues among multiple read-write nodes, problems still exist, such as slow persistence speed, high latency in maintaining cache directories, and timestamp bottlenecks. To address these issues, this study proposes a shared cache architecture-based solution that is combined with novel persistent memory hardware, to realize a three-layer shared architecture database—TampoDB, which includes memory, persistent memory, and storage layers. The transaction execution process was redesigned based on this architecture with optimized timestamps and directories, thereby resolving the aforementioned problems. Experimental results show that TampoDB effectively enhances the persistence speed of transactions.

Key words: cloud-native database, persistent memory, shared cache

中图分类号: