Home > Research > Publications & Outputs > Workload estimation for improving resource mana...

Links

Text available via DOI:

View graph of relations

Workload estimation for improving resource management decisions in the cloud

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Workload estimation for improving resource management decisions in the cloud. / Patel, Jemishkumar ; Jindal, Vasu ; Yen, I-Ling ; Bastani, Farokh ; Xu, Jie; Garraghan, Peter.

2015 IEEE Twelfth International Symposium on Autonomous Decentralized Systems. IEEE, 2015. p. 25-32.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Patel, J, Jindal, V, Yen, I-L, Bastani, F, Xu, J & Garraghan, P 2015, Workload estimation for improving resource management decisions in the cloud. in 2015 IEEE Twelfth International Symposium on Autonomous Decentralized Systems. IEEE, pp. 25-32. https://doi.org/10.1109/ISADS.2015.17

APA

Patel, J., Jindal, V., Yen, I-L., Bastani, F., Xu, J., & Garraghan, P. (2015). Workload estimation for improving resource management decisions in the cloud. In 2015 IEEE Twelfth International Symposium on Autonomous Decentralized Systems (pp. 25-32). IEEE. https://doi.org/10.1109/ISADS.2015.17

Vancouver

Patel J, Jindal V, Yen I-L, Bastani F, Xu J, Garraghan P. Workload estimation for improving resource management decisions in the cloud. In 2015 IEEE Twelfth International Symposium on Autonomous Decentralized Systems. IEEE. 2015. p. 25-32 https://doi.org/10.1109/ISADS.2015.17

Author

Patel, Jemishkumar ; Jindal, Vasu ; Yen, I-Ling ; Bastani, Farokh ; Xu, Jie ; Garraghan, Peter. / Workload estimation for improving resource management decisions in the cloud. 2015 IEEE Twelfth International Symposium on Autonomous Decentralized Systems. IEEE, 2015. pp. 25-32

Bibtex

@inproceedings{1619428075b147d49b2976322420493c,
title = "Workload estimation for improving resource management decisions in the cloud",
abstract = "In cloud computing, good resource management can benefit both cloud users as well as cloud providers. Workload prediction is a crucial step towards achieving good resource management. While it is possible to estimate the workloads of long-running tasks based on the periodicity in their historical workloads, it is difficult to do so for tasks which do not have such recurring workload patterns. In this paper, we present an innovative clustering based resource estimation approach which groups tasks that have similar characteristics into the same cluster. The historical workload data for tasks in a cluster are used to estimate the resources needed by new tasks based on the cluster(s) to which they belong. In particular, for a new task T, we measure T's initial workload and predict to which cluster(s) it may belong. Then, the workload information of the cluster(s) is used to estimate the workload of T. The approach is experimentally evaluated using Google dataset, including resource usage data of over half a million tasks. We develop a workload model based on the dataset which is then used to estimate the workload patterns of several randomly selected tasks from the trace log. The results confirm the effectiveness of this cluster-based method for estimating the resources required by each task.",
keywords = "Estimation, Servers, Google, Resource management, Cloud computing, Clustering algorithms, Time series analysis",
author = "Jemishkumar Patel and Vasu Jindal and I-Ling Yen and Farokh Bastani and Jie Xu and Peter Garraghan",
year = "2015",
month = apr,
day = "29",
doi = "10.1109/ISADS.2015.17",
language = "English",
pages = "25--32",
booktitle = "2015 IEEE Twelfth International Symposium on Autonomous Decentralized Systems",
publisher = "IEEE",

}

RIS

TY - GEN

T1 - Workload estimation for improving resource management decisions in the cloud

AU - Patel, Jemishkumar

AU - Jindal, Vasu

AU - Yen, I-Ling

AU - Bastani, Farokh

AU - Xu, Jie

AU - Garraghan, Peter

PY - 2015/4/29

Y1 - 2015/4/29

N2 - In cloud computing, good resource management can benefit both cloud users as well as cloud providers. Workload prediction is a crucial step towards achieving good resource management. While it is possible to estimate the workloads of long-running tasks based on the periodicity in their historical workloads, it is difficult to do so for tasks which do not have such recurring workload patterns. In this paper, we present an innovative clustering based resource estimation approach which groups tasks that have similar characteristics into the same cluster. The historical workload data for tasks in a cluster are used to estimate the resources needed by new tasks based on the cluster(s) to which they belong. In particular, for a new task T, we measure T's initial workload and predict to which cluster(s) it may belong. Then, the workload information of the cluster(s) is used to estimate the workload of T. The approach is experimentally evaluated using Google dataset, including resource usage data of over half a million tasks. We develop a workload model based on the dataset which is then used to estimate the workload patterns of several randomly selected tasks from the trace log. The results confirm the effectiveness of this cluster-based method for estimating the resources required by each task.

AB - In cloud computing, good resource management can benefit both cloud users as well as cloud providers. Workload prediction is a crucial step towards achieving good resource management. While it is possible to estimate the workloads of long-running tasks based on the periodicity in their historical workloads, it is difficult to do so for tasks which do not have such recurring workload patterns. In this paper, we present an innovative clustering based resource estimation approach which groups tasks that have similar characteristics into the same cluster. The historical workload data for tasks in a cluster are used to estimate the resources needed by new tasks based on the cluster(s) to which they belong. In particular, for a new task T, we measure T's initial workload and predict to which cluster(s) it may belong. Then, the workload information of the cluster(s) is used to estimate the workload of T. The approach is experimentally evaluated using Google dataset, including resource usage data of over half a million tasks. We develop a workload model based on the dataset which is then used to estimate the workload patterns of several randomly selected tasks from the trace log. The results confirm the effectiveness of this cluster-based method for estimating the resources required by each task.

KW - Estimation

KW - Servers

KW - Google

KW - Resource management

KW - Cloud computing

KW - Clustering algorithms

KW - Time series analysis

U2 - 10.1109/ISADS.2015.17

DO - 10.1109/ISADS.2015.17

M3 - Conference contribution/Paper

SP - 25

EP - 32

BT - 2015 IEEE Twelfth International Symposium on Autonomous Decentralized Systems

PB - IEEE

ER -