Virtual-Memory Assisted Buffer Management

Author:

Leis Viktor1ORCID,Alhomssi Adnan2ORCID,Ziegler Tobias3ORCID,Loeck Yannick4ORCID,Dietrich Christian4ORCID

Affiliation:

1. Technische Universität München, München, Germany

2. Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany

3. Technische Universität Darmstadt, Darmstadt, Germany

4. Technische Universität Hamburg, Hamburg, Germany

Abstract

Most database management systems cache pages from storage in a main memory buffer pool. To do this, they either rely on a hash table that translates page identifiers into pointers, or on pointer swizzling which avoids this translation. In this work, we propose vmcache, a buffer manager design that instead uses hardware-supported virtual memory to translate page identifiers to virtual memory addresses. In contrast to existing mmap-based approaches, the DBMS retains control over page faulting and eviction. Our design is portable across modern operating systems, supports arbitrary graph data, enables variable-sized pages, and is easy to implement. One downside of relying on virtual memory is that with fast storage devices the existing operating system primitives for manipulating the page table can become a performance bottleneck. As a second contribution, we therefore propose exmap, which implements scalable page table manipulation on Linux. Together, vmcache and exmap provide flexible, efficient, and scalable buffer management on multi-core CPUs and fast storage devices.

Publisher

Association for Computing Machinery (ACM)

Reference43 articles.

1. 2022. LeanStore - A High-Performance Storage Engine for Modern Hardware. https://leanstore.io/. 2022. LeanStore - A High-Performance Storage Engine for Modern Hardware. https://leanstore.io/.

2. 2022. Lightning Memory-Mapped Database Manager (LMDB). http://www.lmdb.tech/doc/. 2022. Lightning Memory-Mapped Database Manager (LMDB). http://www.lmdb.tech/doc/.

3. 2022. MxKernel - A Bare-Metal Runtime System for Database Operations on Heterogeneous Many-Core Hardware. https://ess.cs.uos.de/research/projects/MxKernel/. 2022. MxKernel - A Bare-Metal Runtime System for Database Operations on Heterogeneous Many-Core Hardware. https://ess.cs.uos.de/research/projects/MxKernel/.

4. 2022. Samsung PCIe Gen 4-enabled PM1733 SSD. https://semiconductor.samsung.com/ssd/enterprise-ssd/pm1733-pm1735/mzwlj3t8hbls-00007/. 2022. Samsung PCIe Gen 4-enabled PM1733 SSD. https://semiconductor.samsung.com/ssd/enterprise-ssd/pm1733-pm1735/mzwlj3t8hbls-00007/.

5. 2022. WiredTiger Storage Engine. https://docs.mongodb.com/manual/core/wiredtiger/. 2022. WiredTiger Storage Engine. https://docs.mongodb.com/manual/core/wiredtiger/.

Cited by 3 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. HPCache: memory-efficient OLAP through proportional caching revisited;The VLDB Journal;2023-12-22

2. Morsels: Explicit Virtual Memory Objects;Proceedings of the 1st Workshop on Disruptive Memory Systems;2023-10-23

3. Evaluation and Refinement of an Explicit Virtual-Memory Primitive;IEEE Access;2023

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