An interface to implement NUMA policies in the Xen hypervisor

被引：11

作者：

Voron, Gauthier ^{[1
]}

Thomas, Gael ^{[2
]}

Quema, Vivien ^{[3
]}

Sens, Pierre ^{[1
]}

机构：

[1] Sorbonne Univ, UPMC Univ Paris 06, INRIA, CNRS,LIP6, Paris, France

[2] Univ Paris Saclay, Telecom SudParis, Gif Sur Yvette, France

[3] Univ Grenoble Alpes, LIG Lab, Grenoble INP, Grenoble, France

来源：

PROCEEDINGS OF THE TWELFTH EUROPEAN CONFERENCE ON COMPUTER SYSTEMS (EUROSYS 2017) | 2017年

关键词：

PERFORMANCE; MEMORY;

D O I：

10.1145/3064176.3064196

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

While virtualization only introduces a small overhead on machines with few cores, this is not the case on larger ones. Most of the overhead on the latter machines is caused by the Non-Uniform Memory Access (NUMA) architecture they are using. In order to reduce this overhead, this paper shows how NUMA placement heuristics can be implemented inside Xen. With an evaluation of 29 applications on a 48-core machine, we show that the NUMA placement heuristics can multiply the performance of 9 applications by more than 2.

引用

页码：453 / 467

页数：15

共 41 条

[1] A comparison of software and hardware techniques for x86 virtualization [J].

Adams, Keith ;

Agesen, Ole .

ACM SIGPLAN NOTICES, 2006, 41 (11) :2-13

[2] Micro-sliced Virtual Processors to Hide the Effect of Discontinuous CPU Availability for Consolidated Systems [J].

Ahn, Jeongseob ;

Park, Chang Hyun ;

Huh, Jaehyuk .

2014 47TH ANNUAL IEEE/ACM INTERNATIONAL SYMPOSIUM ON MICROARCHITECTURE (MICRO), 2014, :394-405

[3]

Aigner M, 2015, ACM SIGPLAN NOTICES, V50, P451, DOI [10.1145/2814270.2814294, 10.1145/2858965.2814294]

[4]

[Anonymous], 2010, P 1 ACM S CLOUD COMP, DOI DOI 10.1145/1807128.1807152

[5]

Boyd-Wickizer Silas., 2010, P 9 USENIX C OPERATI, P1

[6]

Bugnion E, 1997, ACM T COMPUT SYST, V15, P412, DOI [10.1145/265924.265930, 10.1145/269005.266672]

[7]

Cherkasova L, 2005, USENIX ASSOCIATION PROCEEDINGS OF THE GENERAL TRACK: 2005 UNENIX ANNUAL TECHNICAL CONFERENCE, P387

[8] Traffic Management: A Holistic Approach to Memory Placement on NUMA Systems [J].

Dashti, Mohammad ;

Fedorova, Alexandra ;

Funston, Justin ;

Gaud, Fabien ;

Lachaize, Renaud ;

Lepers, Baptiste ;

Quema, Vivien ;

Roth, Mark .

ACM SIGPLAN NOTICES, 2013, 48 (04) :381-393

[9]

David F., 2014, P C OBJ OR PROGR SYS

[10] Everything You Always Wanted to Know About Synchronization but Were Afraid to Ask [J].

David, Tudor ;

Guerraoui, Rachid ;

Trigonakis, Vasileios .

SOSP'13: PROCEEDINGS OF THE TWENTY-FOURTH ACM SYMPOSIUM ON OPERATING SYSTEMS PRINCIPLES, 2013, :33-48

← 1 2 3 4 5 →