Industrial-Strength OLTP Using Main Memory and Many Cores

Hillel Avni; Alisher Aliev; Oren Amor; Aharon Avitzur; Ilan Bronshtein; Eli Ginot; Shay Goikhman; Eliezer Levy; Idan Levy; Fuyang Lu; Liran Mishali; Yeqin Mo; Nir Pachter; Dima Sivov; Vinoth Veeraraghavan; Vladi Vexler; Lei Wang; Peng Wang

doi:10.14778/3415478.3415537

Industrial-Strength OLTP Using Main Memory and Many Cores

Hillel Avni, Alisher Aliev, Oren Amor, Aharon Avitzur, Ilan Bronshtein, Eli Ginot, Shay Goikhman, Eliezer Levy, Idan Levy, Fuyang Lu, Liran Mishali, Yeqin Mo, Nir Pachter, Dima Sivov, Vinoth Veeraraghavan, Vladi Vexler, Lei Wang, Peng Wang

Research output: Contribution to journal › Article › peer-review

13 Scopus citations

Abstract

GaussDB, and its open source version named openGauss, are Huawei’s relational database management systems (RDBMS), featuring a primary disk-based storage engine. This paper presents a new storage engine for GaussDB that is optimized for main memory and many cores. We started from a research prototype which exploits the power of the hardware but is not useful for customers. This paper describes the details of turning this prototype to an industrial storage engine, including integration with GaussDB. Standard benchmarks show that the new engine provides more than 2.5x performance improvement to GaussDB for full TPC-C on Intel’s x86 many-cores servers, as well as on Huawei TaiShan servers powered by ARM64-based Kunpeng CPUs.

Original language	English
Pages (from-to)	3099-3111
Number of pages	13
Journal	Proceedings of the VLDB Endowment
Volume	13
Issue number	12
DOIs	https://doi.org/10.14778/3415478.3415537
State	Published - 1 Jan 2020
Externally published	Yes

ASJC Scopus subject areas

Computer Science (miscellaneous)
General Computer Science

Access to Document

10.14778/3415478.3415537

Cite this

Avni, H., Aliev, A., Amor, O., Avitzur, A., Bronshtein, I., Ginot, E., Goikhman, S., Levy, E., Levy, I., Lu, F., Mishali, L., Mo, Y., Pachter, N., Sivov, D., Veeraraghavan, V., Vexler, V., Wang, L., & Wang, P. (2020). Industrial-Strength OLTP Using Main Memory and Many Cores. Proceedings of the VLDB Endowment, 13(12), 3099-3111. https://doi.org/10.14778/3415478.3415537

@article{b81f357eaa7444d3aad854366501e508,

title = "Industrial-Strength OLTP Using Main Memory and Many Cores",

abstract = "GaussDB, and its open source version named openGauss, are Huawei{\textquoteright}s relational database management systems (RDBMS), featuring a primary disk-based storage engine. This paper presents a new storage engine for GaussDB that is optimized for main memory and many cores. We started from a research prototype which exploits the power of the hardware but is not useful for customers. This paper describes the details of turning this prototype to an industrial storage engine, including integration with GaussDB. Standard benchmarks show that the new engine provides more than 2.5x performance improvement to GaussDB for full TPC-C on Intel{\textquoteright}s x86 many-cores servers, as well as on Huawei TaiShan servers powered by ARM64-based Kunpeng CPUs.",

author = "Hillel Avni and Alisher Aliev and Oren Amor and Aharon Avitzur and Ilan Bronshtein and Eli Ginot and Shay Goikhman and Eliezer Levy and Idan Levy and Fuyang Lu and Liran Mishali and Yeqin Mo and Nir Pachter and Dima Sivov and Vinoth Veeraraghavan and Vladi Vexler and Lei Wang and Peng Wang",

year = "2020",

month = jan,

day = "1",

doi = "10.14778/3415478.3415537",

language = "English",

volume = "13",

pages = "3099--3111",

journal = "Proceedings of the VLDB Endowment",

issn = "2150-8097",

publisher = "Very Large Data Base Endowment Inc.",

number = "12",

}

TY - JOUR

T1 - Industrial-Strength OLTP Using Main Memory and Many Cores

AU - Avni, Hillel

AU - Aliev, Alisher

AU - Amor, Oren

AU - Avitzur, Aharon

AU - Bronshtein, Ilan

AU - Ginot, Eli

AU - Goikhman, Shay

AU - Levy, Eliezer

AU - Levy, Idan

AU - Lu, Fuyang

AU - Mishali, Liran

AU - Mo, Yeqin

AU - Pachter, Nir

AU - Sivov, Dima

AU - Veeraraghavan, Vinoth

AU - Vexler, Vladi

AU - Wang, Lei

AU - Wang, Peng

PY - 2020/1/1

Y1 - 2020/1/1

N2 - GaussDB, and its open source version named openGauss, are Huawei’s relational database management systems (RDBMS), featuring a primary disk-based storage engine. This paper presents a new storage engine for GaussDB that is optimized for main memory and many cores. We started from a research prototype which exploits the power of the hardware but is not useful for customers. This paper describes the details of turning this prototype to an industrial storage engine, including integration with GaussDB. Standard benchmarks show that the new engine provides more than 2.5x performance improvement to GaussDB for full TPC-C on Intel’s x86 many-cores servers, as well as on Huawei TaiShan servers powered by ARM64-based Kunpeng CPUs.

AB - GaussDB, and its open source version named openGauss, are Huawei’s relational database management systems (RDBMS), featuring a primary disk-based storage engine. This paper presents a new storage engine for GaussDB that is optimized for main memory and many cores. We started from a research prototype which exploits the power of the hardware but is not useful for customers. This paper describes the details of turning this prototype to an industrial storage engine, including integration with GaussDB. Standard benchmarks show that the new engine provides more than 2.5x performance improvement to GaussDB for full TPC-C on Intel’s x86 many-cores servers, as well as on Huawei TaiShan servers powered by ARM64-based Kunpeng CPUs.

UR - http://www.scopus.com/inward/record.url?scp=85103301087&partnerID=8YFLogxK

U2 - 10.14778/3415478.3415537

DO - 10.14778/3415478.3415537

M3 - Article

AN - SCOPUS:85103301087

SN - 2150-8097

VL - 13

SP - 3099

EP - 3111

JO - Proceedings of the VLDB Endowment

JF - Proceedings of the VLDB Endowment

IS - 12

ER -

Industrial-Strength OLTP Using Main Memory and Many Cores

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this