Towards a leaner geo-distributed cloud infrastructure

Iyswarya Narayanan; Aman Kansal; Anand Sivasubramaniam; Bhuvan Urgaonkar; Sriram Govindan

Towards a leaner geo-distributed cloud infrastructure

Iyswarya Narayanan, Aman Kansal, Anand Sivasubramaniam, Bhuvan Urgaonkar, Sriram Govindan

Research output: Contribution to conference › Paper › peer-review

Abstract

Modern cloud infrastructures are geo-distributed. Geo-distribution offers many advantages but can increase the total cloud capacity required. To achieve low latency, geo-distribution forfeits statistical multiplexing of demand that a single data center could benefit from. Geo-distribution also complicates software design due to storage consistency issues. On the other hand, geo-distribution can lower costs through eliminating redundancies at individual sites or exploiting regional differences in energy prices. We discuss several factors that influence geo-distributed capacity provisioning, and quantify latency, availability, and capacity trade-offs that emerge. We describe open research challenges in designing software that efficiently uses cloud capacity.

Original language	English (US)
State	Published - 2014
Event	6th USENIX Workshop on Hot Topics in Cloud Computing, HotCloud 2014 - Philadelphia, United States Duration: Jun 17 2014 → Jun 18 2014

Conference

Conference	6th USENIX Workshop on Hot Topics in Cloud Computing, HotCloud 2014
Country/Territory	United States
City	Philadelphia
Period	6/17/14 → 6/18/14

All Science Journal Classification (ASJC) codes

Computer Networks and Communications
Software

Cite this

@conference{8b44e6a16b3746048cad230682adb7e7,

title = "Towards a leaner geo-distributed cloud infrastructure",

abstract = "Modern cloud infrastructures are geo-distributed. Geo-distribution offers many advantages but can increase the total cloud capacity required. To achieve low latency, geo-distribution forfeits statistical multiplexing of demand that a single data center could benefit from. Geo-distribution also complicates software design due to storage consistency issues. On the other hand, geo-distribution can lower costs through eliminating redundancies at individual sites or exploiting regional differences in energy prices. We discuss several factors that influence geo-distributed capacity provisioning, and quantify latency, availability, and capacity trade-offs that emerge. We describe open research challenges in designing software that efficiently uses cloud capacity.",

author = "Iyswarya Narayanan and Aman Kansal and Anand Sivasubramaniam and Bhuvan Urgaonkar and Sriram Govindan",

note = "Publisher Copyright: {\textcopyright} 2014 USENIX Association. All rights reserved.; 6th USENIX Workshop on Hot Topics in Cloud Computing, HotCloud 2014 ; Conference date: 17-06-2014 Through 18-06-2014",

year = "2014",

language = "English (US)",

}

TY - CONF

T1 - Towards a leaner geo-distributed cloud infrastructure

AU - Narayanan, Iyswarya

AU - Kansal, Aman

AU - Sivasubramaniam, Anand

AU - Urgaonkar, Bhuvan

AU - Govindan, Sriram

PY - 2014

Y1 - 2014

N2 - Modern cloud infrastructures are geo-distributed. Geo-distribution offers many advantages but can increase the total cloud capacity required. To achieve low latency, geo-distribution forfeits statistical multiplexing of demand that a single data center could benefit from. Geo-distribution also complicates software design due to storage consistency issues. On the other hand, geo-distribution can lower costs through eliminating redundancies at individual sites or exploiting regional differences in energy prices. We discuss several factors that influence geo-distributed capacity provisioning, and quantify latency, availability, and capacity trade-offs that emerge. We describe open research challenges in designing software that efficiently uses cloud capacity.

AB - Modern cloud infrastructures are geo-distributed. Geo-distribution offers many advantages but can increase the total cloud capacity required. To achieve low latency, geo-distribution forfeits statistical multiplexing of demand that a single data center could benefit from. Geo-distribution also complicates software design due to storage consistency issues. On the other hand, geo-distribution can lower costs through eliminating redundancies at individual sites or exploiting regional differences in energy prices. We discuss several factors that influence geo-distributed capacity provisioning, and quantify latency, availability, and capacity trade-offs that emerge. We describe open research challenges in designing software that efficiently uses cloud capacity.

UR - http://www.scopus.com/inward/record.url?scp=84992109948&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84992109948&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:84992109948

T2 - 6th USENIX Workshop on Hot Topics in Cloud Computing, HotCloud 2014

Y2 - 17 June 2014 through 18 June 2014

ER -

Towards a leaner geo-distributed cloud infrastructure

Abstract

Conference

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this