Impacts of task placement and bandwidth allocation on stream analytics

Walid A.Y. Aljoby; Tom Z.J. Fu; Richard T.B. Ma

doi:10.1109/ICNP.2017.8117589

Impacts of task placement and bandwidth allocation on stream analytics

Walid A.Y. Aljoby, Tom Z.J. Fu, Richard T.B. Ma

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

5 Scopus citations

Abstract

We consider data intensive cloud-based stream analytics where data transmission through the underlying communication network is the cause of the performance bottleneck. Two key inter-related problems are investigated: task placement and bandwidth allocation. We seek to answer the following questions. How does task placement make impact on the application-level throughput? Does a careful bandwidth allocation among data flows traversing a bottleneck link results in better performance? In this paper, we address these questions by conducting measurement-driven analysis in a SDN-enabled computer cluster running stream processing applications on top of Apache Storm. The results reveal (i) how tasks are assigned to computing nodes make large difference in application level performance; (ii) under certain task placement, a proper bandwidth allocation helps further improve the performance as compared to the default TCP mechanism; and (iii) task placement and bandwidth allocation are collaboratively making effects in overall performance.

Original language	English
Title of host publication	2017 IEEE 25th International Conference on Network Protocols, ICNP 2017
Publisher	IEEE Computer Society
ISBN (Electronic)	9781509065011
DOIs	https://doi.org/10.1109/ICNP.2017.8117589
State	Published - 21 Nov 2017
Externally published	Yes
Event	25th IEEE International Conference on Network Protocols, ICNP 2017 - Toronto, Canada Duration: 10 Oct 2017 → 13 Oct 2017

Publication series

Name	Proceedings - International Conference on Network Protocols, ICNP
Volume	2017-October
ISSN (Print)	1092-1648

Conference

Conference	25th IEEE International Conference on Network Protocols, ICNP 2017
Country/Territory	Canada
City	Toronto
Period	10/10/17 → 13/10/17

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

ASJC Scopus subject areas

Computer Networks and Communications
Software

Access to Document

10.1109/ICNP.2017.8117589

Cite this

Aljoby, W. A. Y., Fu, T. Z. J., & Ma, R. T. B. (2017). Impacts of task placement and bandwidth allocation on stream analytics. In 2017 IEEE 25th International Conference on Network Protocols, ICNP 2017 Article 8117589 (Proceedings - International Conference on Network Protocols, ICNP; Vol. 2017-October). IEEE Computer Society. https://doi.org/10.1109/ICNP.2017.8117589

@inproceedings{424e16f3fb364a1abb5977770ce6e3fb,

title = "Impacts of task placement and bandwidth allocation on stream analytics",

abstract = "We consider data intensive cloud-based stream analytics where data transmission through the underlying communication network is the cause of the performance bottleneck. Two key inter-related problems are investigated: task placement and bandwidth allocation. We seek to answer the following questions. How does task placement make impact on the application-level throughput? Does a careful bandwidth allocation among data flows traversing a bottleneck link results in better performance? In this paper, we address these questions by conducting measurement-driven analysis in a SDN-enabled computer cluster running stream processing applications on top of Apache Storm. The results reveal (i) how tasks are assigned to computing nodes make large difference in application level performance; (ii) under certain task placement, a proper bandwidth allocation helps further improve the performance as compared to the default TCP mechanism; and (iii) task placement and bandwidth allocation are collaboratively making effects in overall performance.",

author = "Aljoby, \{Walid A.Y.\} and Fu, \{Tom Z.J.\} and Ma, \{Richard T.B.\}",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 25th IEEE International Conference on Network Protocols, ICNP 2017 ; Conference date: 10-10-2017 Through 13-10-2017",

year = "2017",

month = nov,

day = "21",

doi = "10.1109/ICNP.2017.8117589",

language = "English",

series = "Proceedings - International Conference on Network Protocols, ICNP",

publisher = "IEEE Computer Society",

booktitle = "2017 IEEE 25th International Conference on Network Protocols, ICNP 2017",

address = "United States",

}

Aljoby, WAY, Fu, TZJ & Ma, RTB 2017, Impacts of task placement and bandwidth allocation on stream analytics. in 2017 IEEE 25th International Conference on Network Protocols, ICNP 2017., 8117589, Proceedings - International Conference on Network Protocols, ICNP, vol. 2017-October, IEEE Computer Society, 25th IEEE International Conference on Network Protocols, ICNP 2017, Toronto, Canada, 10/10/17. https://doi.org/10.1109/ICNP.2017.8117589

Impacts of task placement and bandwidth allocation on stream analytics. / Aljoby, Walid A.Y.; Fu, Tom Z.J.; Ma, Richard T.B.
2017 IEEE 25th International Conference on Network Protocols, ICNP 2017. IEEE Computer Society, 2017. 8117589 (Proceedings - International Conference on Network Protocols, ICNP; Vol. 2017-October).

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

TY - GEN

T1 - Impacts of task placement and bandwidth allocation on stream analytics

AU - Aljoby, Walid A.Y.

AU - Fu, Tom Z.J.

AU - Ma, Richard T.B.

PY - 2017/11/21

Y1 - 2017/11/21

N2 - We consider data intensive cloud-based stream analytics where data transmission through the underlying communication network is the cause of the performance bottleneck. Two key inter-related problems are investigated: task placement and bandwidth allocation. We seek to answer the following questions. How does task placement make impact on the application-level throughput? Does a careful bandwidth allocation among data flows traversing a bottleneck link results in better performance? In this paper, we address these questions by conducting measurement-driven analysis in a SDN-enabled computer cluster running stream processing applications on top of Apache Storm. The results reveal (i) how tasks are assigned to computing nodes make large difference in application level performance; (ii) under certain task placement, a proper bandwidth allocation helps further improve the performance as compared to the default TCP mechanism; and (iii) task placement and bandwidth allocation are collaboratively making effects in overall performance.

AB - We consider data intensive cloud-based stream analytics where data transmission through the underlying communication network is the cause of the performance bottleneck. Two key inter-related problems are investigated: task placement and bandwidth allocation. We seek to answer the following questions. How does task placement make impact on the application-level throughput? Does a careful bandwidth allocation among data flows traversing a bottleneck link results in better performance? In this paper, we address these questions by conducting measurement-driven analysis in a SDN-enabled computer cluster running stream processing applications on top of Apache Storm. The results reveal (i) how tasks are assigned to computing nodes make large difference in application level performance; (ii) under certain task placement, a proper bandwidth allocation helps further improve the performance as compared to the default TCP mechanism; and (iii) task placement and bandwidth allocation are collaboratively making effects in overall performance.

UR - https://www.scopus.com/pages/publications/85041420162

U2 - 10.1109/ICNP.2017.8117589

DO - 10.1109/ICNP.2017.8117589

M3 - Conference contribution

AN - SCOPUS:85041420162

T3 - Proceedings - International Conference on Network Protocols, ICNP

BT - 2017 IEEE 25th International Conference on Network Protocols, ICNP 2017

PB - IEEE Computer Society

T2 - 25th IEEE International Conference on Network Protocols, ICNP 2017

Y2 - 10 October 2017 through 13 October 2017

ER -

Impacts of task placement and bandwidth allocation on stream analytics

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this