NVM: Is it Not Very Meaningful for Databases?

Author:

Koutsoukos Dimitrios1,Bhartia Raghav1,Friedman Michal1,Klimovic Ana1,Alonso Gustavo1

Affiliation:

1. ETH Zurich, Switzerland

Abstract

Persistent or Non Volatile Memory (PMEM) offers expanded memory capacity and faster access to persistent storage. However, there is no comprehensive empirical analysis of existing database engines under different PMEM modes, to understand how databases can benefit from the various hardware configurations. To this end, we analyze multiple different engines under common benchmarks with PMEM in AppDirect mode and Memory mode. Our results show that PMEM in Memory mode does not offer any clear performance advantage despite the larger volatile memory capacity. Also, using PMEM as persistent storage usually speeds up query execution, but with some caveats as the I/O path is not fully optimized and therefore does not always justify the additional cost. We show this to be the case through a comprehensive evaluation of different engines and database configurations under different workloads.

Publisher

Association for Computing Machinery (ACM)

Subject

General Earth and Planetary Sciences,Water Science and Technology,Geography, Planning and Development

Reference65 articles.

1. Accessed 2021-10-18. Peloton. https://pelotondb.io/. Accessed 2021-10-18. Peloton. https://pelotondb.io/.

2. Accessed 2021-10-19. Intel®Optane©DC Persistent Memory. https://www.intel.com/content/www/us/en/architecture-and-technology/optane-dc-persistent-memory.html. Accessed 2021-10-19. Intel®Optane©DC Persistent Memory. https://www.intel.com/content/www/us/en/architecture-and-technology/optane-dc-persistent-memory.html.

3. Accessed 2021-10-25. Intel Vtune Platform Profiler. https://www.intel.com/content/dam/develop/external/us/en/documents/vtuneplatformprofilerwhitepaper.pdf. Accessed 2021-10-25. Intel Vtune Platform Profiler. https://www.intel.com/content/dam/develop/external/us/en/documents/vtuneplatformprofilerwhitepaper.pdf.

4. Accessed 2021-11-01. Faster access to more data. https://www.intel.ca/content/www/ca/en/architecture-and-technology/optane-technology/faster-access-to-more-data-article-brief.html. Accessed 2021-11-01. Faster access to more data. https://www.intel.ca/content/www/ca/en/architecture-and-technology/optane-technology/faster-access-to-more-data-article-brief.html.

5. Accessed 2021-11-02. InnoDB: Parallel read of index. https://dev.mysql.com/worklog/task/?id=11720#tabs-11720-2. Accessed 2021-11-02. InnoDB: Parallel read of index. https://dev.mysql.com/worklog/task/?id=11720#tabs-11720-2.

Cited by 2 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Sorting on Byte-Addressable Storage: The Resurgence of Tree Structure;Proceedings of the VLDB Endowment;2024-02

2. An NVM Performance Study Towards Whole System Persistence on Server Platforms;Proceedings of the 1st Workshop on Disruptive Memory Systems;2023-10-23

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3