TailCutter: Wisely cutting tail latency in cloud CDN under cost constraints

Zeqi Lai; Yong Cui; Minming Li; Zhenhua Li; Ningwei Dai; Yuchi Chen

doi:10.1109/INFOCOM.2016.7524535

TailCutter: Wisely cutting tail latency in cloud CDN under cost constraints

Zeqi Lai^*, Yong Cui, Minming Li, Zhenhua Li, Ningwei Dai, Yuchi Chen

^*Corresponding author for this work

Department of Computer Science

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

15 Citations (Scopus)

Abstract

Cloud computing platforms enable applications to offer low latency access to user data by offering storage services in several geographically distributed data centers. In this paper, we identify the high tail latency problem in cloud CDN via analyzing a large-scale dataset collected from 783,944 users in a major cloud CDN. We find that the data downloading latency in cloud CDN is highly variable, which may significantly degrade the user experience of applications. To address the problem, we present TailCutter, a workload scheduling mechanism that aims at optimizing the tail latency while meeting the cost constraint given by application providers. We further design the Maximum Tail Minimization Algorithm (MTMA) working in TailCutter mechanism to optimally solve the Tail Latency Minimization (TLM) problem in polynomial time. We implement TailCutter across data centers of Amazon S3 and Microsoft Azure. Our extensive evaluation using large-scale real world data traces shows that TailCutter can reduce up to 68% 99th percentile user-perceived latency in comparison with alternative solutions under cost constraints.

© 2016 IEEE

Original language	English
Title of host publication	IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications
Publisher	IEEE
ISBN (Electronic)	978-1-4673-9953-1
DOIs	https://doi.org/10.1109/INFOCOM.2016.7524535
Publication status	Published - 2016
Event	35th Annual IEEE International Conference on Computer Communications, IEEE INFOCOM 2016 - San Francisco, United States Duration: 10 Apr 2016 → 14 Apr 2016 http://infocom2016.ieee-infocom.org/

Publication series

Name
Volume	2016-July
ISSN (Print)	0743-166X

Conference

Conference	35th Annual IEEE International Conference on Computer Communications, IEEE INFOCOM 2016
Place	United States
City	San Francisco
Period	10/04/16 → 14/04/16
Internet address	http://infocom2016.ieee-infocom.org/

Access Output

10.1109/INFOCOM.2016.7524535

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{55d1485521fb4788b5ef0ddabc31daa6,

title = "TailCutter: Wisely cutting tail latency in cloud CDN under cost constraints",

abstract = "Cloud computing platforms enable applications to offer low latency access to user data by offering storage services in several geographically distributed data centers. In this paper, we identify the high tail latency problem in cloud CDN via analyzing a large-scale dataset collected from 783,944 users in a major cloud CDN. We find that the data downloading latency in cloud CDN is highly variable, which may significantly degrade the user experience of applications. To address the problem, we present TailCutter, a workload scheduling mechanism that aims at optimizing the tail latency while meeting the cost constraint given by application providers. We further design the Maximum Tail Minimization Algorithm (MTMA) working in TailCutter mechanism to optimally solve the Tail Latency Minimization (TLM) problem in polynomial time. We implement TailCutter across data centers of Amazon S3 and Microsoft Azure. Our extensive evaluation using large-scale real world data traces shows that TailCutter can reduce up to 68% 99th percentile user-perceived latency in comparison with alternative solutions under cost constraints.{\textcopyright} 2016 IEEE",

author = "Zeqi Lai and Yong Cui and Minming Li and Zhenhua Li and Ningwei Dai and Yuchi Chen",

year = "2016",

doi = "10.1109/INFOCOM.2016.7524535",

language = "English",

publisher = "IEEE",

booktitle = "IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications",

address = "United States",

note = "35th Annual IEEE International Conference on Computer Communications, IEEE INFOCOM 2016 ; Conference date: 10-04-2016 Through 14-04-2016",

url = "http://infocom2016.ieee-infocom.org/",

}

Lai, Z, Cui, Y, Li, M, Li, Z, Dai, N & Chen, Y 2016, TailCutter: Wisely cutting tail latency in cloud CDN under cost constraints. in IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications., 7524535, IEEE, 35th Annual IEEE International Conference on Computer Communications, IEEE INFOCOM 2016, San Francisco, United States, 10/04/16. https://doi.org/10.1109/INFOCOM.2016.7524535

TailCutter: Wisely cutting tail latency in cloud CDN under cost constraints. / Lai, Zeqi; Cui, Yong; Li, Minming et al.
IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications. IEEE, 2016. 7524535.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - TailCutter

T2 - 35th Annual IEEE International Conference on Computer Communications, IEEE INFOCOM 2016

AU - Lai, Zeqi

AU - Cui, Yong

AU - Li, Minming

AU - Li, Zhenhua

AU - Dai, Ningwei

AU - Chen, Yuchi

PY - 2016

Y1 - 2016

N2 - Cloud computing platforms enable applications to offer low latency access to user data by offering storage services in several geographically distributed data centers. In this paper, we identify the high tail latency problem in cloud CDN via analyzing a large-scale dataset collected from 783,944 users in a major cloud CDN. We find that the data downloading latency in cloud CDN is highly variable, which may significantly degrade the user experience of applications. To address the problem, we present TailCutter, a workload scheduling mechanism that aims at optimizing the tail latency while meeting the cost constraint given by application providers. We further design the Maximum Tail Minimization Algorithm (MTMA) working in TailCutter mechanism to optimally solve the Tail Latency Minimization (TLM) problem in polynomial time. We implement TailCutter across data centers of Amazon S3 and Microsoft Azure. Our extensive evaluation using large-scale real world data traces shows that TailCutter can reduce up to 68% 99th percentile user-perceived latency in comparison with alternative solutions under cost constraints.© 2016 IEEE

AB - Cloud computing platforms enable applications to offer low latency access to user data by offering storage services in several geographically distributed data centers. In this paper, we identify the high tail latency problem in cloud CDN via analyzing a large-scale dataset collected from 783,944 users in a major cloud CDN. We find that the data downloading latency in cloud CDN is highly variable, which may significantly degrade the user experience of applications. To address the problem, we present TailCutter, a workload scheduling mechanism that aims at optimizing the tail latency while meeting the cost constraint given by application providers. We further design the Maximum Tail Minimization Algorithm (MTMA) working in TailCutter mechanism to optimally solve the Tail Latency Minimization (TLM) problem in polynomial time. We implement TailCutter across data centers of Amazon S3 and Microsoft Azure. Our extensive evaluation using large-scale real world data traces shows that TailCutter can reduce up to 68% 99th percentile user-perceived latency in comparison with alternative solutions under cost constraints.© 2016 IEEE

UR - http://www.scopus.com/inward/record.url?scp=84983359025&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-84983359025&origin=recordpage

U2 - 10.1109/INFOCOM.2016.7524535

DO - 10.1109/INFOCOM.2016.7524535

M3 - RGC 32 - Refereed conference paper (with host publication)

BT - IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications

PB - IEEE

Y2 - 10 April 2016 through 14 April 2016

ER -

TailCutter: Wisely cutting tail latency in cloud CDN under cost constraints

Abstract

Publication series

Conference

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this