Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS

Yinjun Wu; Kwanghyun Park; Rathijit Sen; Brian Kroth; Jaeyoung Do

doi:10.1145/3399666.3399898

Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS

Yinjun Wu, Kwanghyun Park, Rathijit Sen, Brian Kroth, Jaeyoung Do

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Non-volatile memory (NVM) is an emerging technology, which has the persistence characteristics of large capacity storage devices, while providing the low access latency and byte-addressablity of traditional DRAM memory. In this paper, we provide extensive performance evaluations on a recently released NVM device, Intel Optane DC Persistent Memory (PMem), under different configurations with several micro-benchmark tools. Further, we evaluate OLTP and OLAP database workloads with Microsoft SQL Server 2019 when using PMem as buffer pool or persistent storage. From the lessons learned we share some recommendations for future DBMS design with PMem, e.g. simple hardware or software changes are not enough for the best use of PMem in DBMSs.

Original language	English
Title of host publication	Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020
Publisher	Association for Computing Machinery
ISBN (Electronic)	9781450380249
DOIs	https://doi.org/10.1145/3399666.3399898
Publication status	Published - 2020 Jun 15
Event	16th International Workshop on Data Management on New Hardware, DaMoN 2020 - Portland, United States Duration: 2020 Jun 15 → …

Publication series

Name	Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020

Conference

Conference	16th International Workshop on Data Management on New Hardware, DaMoN 2020
Country/Territory	United States
City	Portland
Period	20/6/15 → …

Bibliographical note

Publisher Copyright:
© 2020 ACM.

All Science Journal Classification (ASJC) codes

Software
Computer Networks and Communications
Hardware and Architecture

Access to Document

10.1145/3399666.3399898

Cite this

Wu, Y., Park, K., Sen, R., Kroth, B., & Do, J. (2020). Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS. In Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020 (Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020). Association for Computing Machinery. https://doi.org/10.1145/3399666.3399898

Wu, Yinjun ; Park, Kwanghyun ; Sen, Rathijit et al. / Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS. Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020. Association for Computing Machinery, 2020. (Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020).

@inproceedings{8a19e1bd983a44d48f575ae756b8dfa0,

title = "Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS",

abstract = "Non-volatile memory (NVM) is an emerging technology, which has the persistence characteristics of large capacity storage devices, while providing the low access latency and byte-addressablity of traditional DRAM memory. In this paper, we provide extensive performance evaluations on a recently released NVM device, Intel Optane DC Persistent Memory (PMem), under different configurations with several micro-benchmark tools. Further, we evaluate OLTP and OLAP database workloads with Microsoft SQL Server 2019 when using PMem as buffer pool or persistent storage. From the lessons learned we share some recommendations for future DBMS design with PMem, e.g. simple hardware or software changes are not enough for the best use of PMem in DBMSs.",

author = "Yinjun Wu and Kwanghyun Park and Rathijit Sen and Brian Kroth and Jaeyoung Do",

note = "Publisher Copyright: {\textcopyright} 2020 ACM.; 16th International Workshop on Data Management on New Hardware, DaMoN 2020 ; Conference date: 15-06-2020",

year = "2020",

month = jun,

day = "15",

doi = "10.1145/3399666.3399898",

language = "English",

series = "Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020",

publisher = "Association for Computing Machinery",

booktitle = "Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020",

}

Wu, Y, Park, K, Sen, R, Kroth, B & Do, J 2020, Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS. in Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020. Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020, Association for Computing Machinery, 16th International Workshop on Data Management on New Hardware, DaMoN 2020, Portland, United States, 20/6/15. https://doi.org/10.1145/3399666.3399898

Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS. / Wu, Yinjun; Park, Kwanghyun; Sen, Rathijit et al.

Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020. Association for Computing Machinery, 2020. (Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS

AU - Wu, Yinjun

AU - Park, Kwanghyun

AU - Sen, Rathijit

AU - Kroth, Brian

AU - Do, Jaeyoung

PY - 2020/6/15

Y1 - 2020/6/15

N2 - Non-volatile memory (NVM) is an emerging technology, which has the persistence characteristics of large capacity storage devices, while providing the low access latency and byte-addressablity of traditional DRAM memory. In this paper, we provide extensive performance evaluations on a recently released NVM device, Intel Optane DC Persistent Memory (PMem), under different configurations with several micro-benchmark tools. Further, we evaluate OLTP and OLAP database workloads with Microsoft SQL Server 2019 when using PMem as buffer pool or persistent storage. From the lessons learned we share some recommendations for future DBMS design with PMem, e.g. simple hardware or software changes are not enough for the best use of PMem in DBMSs.

AB - Non-volatile memory (NVM) is an emerging technology, which has the persistence characteristics of large capacity storage devices, while providing the low access latency and byte-addressablity of traditional DRAM memory. In this paper, we provide extensive performance evaluations on a recently released NVM device, Intel Optane DC Persistent Memory (PMem), under different configurations with several micro-benchmark tools. Further, we evaluate OLTP and OLAP database workloads with Microsoft SQL Server 2019 when using PMem as buffer pool or persistent storage. From the lessons learned we share some recommendations for future DBMS design with PMem, e.g. simple hardware or software changes are not enough for the best use of PMem in DBMSs.

UR - http://www.scopus.com/inward/record.url?scp=85087654131&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85087654131&partnerID=8YFLogxK

U2 - 10.1145/3399666.3399898

DO - 10.1145/3399666.3399898

M3 - Conference contribution

AN - SCOPUS:85087654131

T3 - Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020

BT - Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020

PB - Association for Computing Machinery

T2 - 16th International Workshop on Data Management on New Hardware, DaMoN 2020

Y2 - 15 June 2020

ER -

Wu Y, Park K, Sen R, Kroth B, Do J. Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS. In Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020. Association for Computing Machinery. 2020. (Proceedings of the 16th International Workshop on Data Management on New Hardware, DaMoN 2020). doi: 10.1145/3399666.3399898

Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this