Scheduling parallel tasks onto opportunistically available cloud resources

Ting He, Shiyao Chen, Hyoil Kim, Lang Tong, Kang Won Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

16 Citations (Scopus)

Abstract

We consider the problem of opportunistically scheduling low-priority tasks onto underutilized computation resources in the cloud left by high-priority tasks. To avoid conflicts with high-priority tasks, the scheduler must suspend the low-priority tasks (causing waiting), or move them to other underutilized servers (causing migration), if the high-priority tasks resume. The goal of opportunistic scheduling is to schedule the low-priority tasks onto intermittently available server resources while minimizing the combined cost of waiting and migration. Moreover, we aim to support multiple parallel low-priority tasks with synchronization constraints. Under the assumption that servers' availability to low-priority tasks can be modeled as ON/OFF Markov chains, we have shown that the optimal solution requires solving a Markov Decision Process (MDP) that has exponential complexity, and efficient solutions are known only in the case of homogeneously behaving servers. In this paper, we propose an efficient heuristic scheduling policy by formulating the problem as restless Multi-Armed Bandits (MAB) under relaxed synchronization. We prove the index ability of the problem and provide closed-form formulas to compute the indices. Our evaluation using real data center traces shows that the performance result closely matches the prediction by the Markov chain model, and the proposed index policy achieves consistently good performance under various server dynamics compared with the existing policies.

Original languageEnglish (US)
Title of host publicationProceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012
Pages180-187
Number of pages8
DOIs
StatePublished - Oct 2 2012
Event2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012 - Honolulu, HI, United States
Duration: Jun 24 2012Jun 29 2012

Publication series

NameProceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012

Other

Other2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012
CountryUnited States
CityHonolulu, HI
Period6/24/126/29/12

Fingerprint

Servers
Scheduling
Markov processes
Synchronization
Availability
Costs

All Science Journal Classification (ASJC) codes

  • Software

Cite this

He, T., Chen, S., Kim, H., Tong, L., & Lee, K. W. (2012). Scheduling parallel tasks onto opportunistically available cloud resources. In Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012 (pp. 180-187). [6253504] (Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012). https://doi.org/10.1109/CLOUD.2012.15
He, Ting ; Chen, Shiyao ; Kim, Hyoil ; Tong, Lang ; Lee, Kang Won. / Scheduling parallel tasks onto opportunistically available cloud resources. Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012. 2012. pp. 180-187 (Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012).
@inproceedings{b14d8aa137304db1984f4b1b7fd3af51,
title = "Scheduling parallel tasks onto opportunistically available cloud resources",
abstract = "We consider the problem of opportunistically scheduling low-priority tasks onto underutilized computation resources in the cloud left by high-priority tasks. To avoid conflicts with high-priority tasks, the scheduler must suspend the low-priority tasks (causing waiting), or move them to other underutilized servers (causing migration), if the high-priority tasks resume. The goal of opportunistic scheduling is to schedule the low-priority tasks onto intermittently available server resources while minimizing the combined cost of waiting and migration. Moreover, we aim to support multiple parallel low-priority tasks with synchronization constraints. Under the assumption that servers' availability to low-priority tasks can be modeled as ON/OFF Markov chains, we have shown that the optimal solution requires solving a Markov Decision Process (MDP) that has exponential complexity, and efficient solutions are known only in the case of homogeneously behaving servers. In this paper, we propose an efficient heuristic scheduling policy by formulating the problem as restless Multi-Armed Bandits (MAB) under relaxed synchronization. We prove the index ability of the problem and provide closed-form formulas to compute the indices. Our evaluation using real data center traces shows that the performance result closely matches the prediction by the Markov chain model, and the proposed index policy achieves consistently good performance under various server dynamics compared with the existing policies.",
author = "Ting He and Shiyao Chen and Hyoil Kim and Lang Tong and Lee, {Kang Won}",
year = "2012",
month = "10",
day = "2",
doi = "10.1109/CLOUD.2012.15",
language = "English (US)",
isbn = "9780769547558",
series = "Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012",
pages = "180--187",
booktitle = "Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012",

}

He, T, Chen, S, Kim, H, Tong, L & Lee, KW 2012, Scheduling parallel tasks onto opportunistically available cloud resources. in Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012., 6253504, Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012, pp. 180-187, 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012, Honolulu, HI, United States, 6/24/12. https://doi.org/10.1109/CLOUD.2012.15

Scheduling parallel tasks onto opportunistically available cloud resources. / He, Ting; Chen, Shiyao; Kim, Hyoil; Tong, Lang; Lee, Kang Won.

Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012. 2012. p. 180-187 6253504 (Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Scheduling parallel tasks onto opportunistically available cloud resources

AU - He, Ting

AU - Chen, Shiyao

AU - Kim, Hyoil

AU - Tong, Lang

AU - Lee, Kang Won

PY - 2012/10/2

Y1 - 2012/10/2

N2 - We consider the problem of opportunistically scheduling low-priority tasks onto underutilized computation resources in the cloud left by high-priority tasks. To avoid conflicts with high-priority tasks, the scheduler must suspend the low-priority tasks (causing waiting), or move them to other underutilized servers (causing migration), if the high-priority tasks resume. The goal of opportunistic scheduling is to schedule the low-priority tasks onto intermittently available server resources while minimizing the combined cost of waiting and migration. Moreover, we aim to support multiple parallel low-priority tasks with synchronization constraints. Under the assumption that servers' availability to low-priority tasks can be modeled as ON/OFF Markov chains, we have shown that the optimal solution requires solving a Markov Decision Process (MDP) that has exponential complexity, and efficient solutions are known only in the case of homogeneously behaving servers. In this paper, we propose an efficient heuristic scheduling policy by formulating the problem as restless Multi-Armed Bandits (MAB) under relaxed synchronization. We prove the index ability of the problem and provide closed-form formulas to compute the indices. Our evaluation using real data center traces shows that the performance result closely matches the prediction by the Markov chain model, and the proposed index policy achieves consistently good performance under various server dynamics compared with the existing policies.

AB - We consider the problem of opportunistically scheduling low-priority tasks onto underutilized computation resources in the cloud left by high-priority tasks. To avoid conflicts with high-priority tasks, the scheduler must suspend the low-priority tasks (causing waiting), or move them to other underutilized servers (causing migration), if the high-priority tasks resume. The goal of opportunistic scheduling is to schedule the low-priority tasks onto intermittently available server resources while minimizing the combined cost of waiting and migration. Moreover, we aim to support multiple parallel low-priority tasks with synchronization constraints. Under the assumption that servers' availability to low-priority tasks can be modeled as ON/OFF Markov chains, we have shown that the optimal solution requires solving a Markov Decision Process (MDP) that has exponential complexity, and efficient solutions are known only in the case of homogeneously behaving servers. In this paper, we propose an efficient heuristic scheduling policy by formulating the problem as restless Multi-Armed Bandits (MAB) under relaxed synchronization. We prove the index ability of the problem and provide closed-form formulas to compute the indices. Our evaluation using real data center traces shows that the performance result closely matches the prediction by the Markov chain model, and the proposed index policy achieves consistently good performance under various server dynamics compared with the existing policies.

UR - http://www.scopus.com/inward/record.url?scp=84866771828&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866771828&partnerID=8YFLogxK

U2 - 10.1109/CLOUD.2012.15

DO - 10.1109/CLOUD.2012.15

M3 - Conference contribution

AN - SCOPUS:84866771828

SN - 9780769547558

T3 - Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012

SP - 180

EP - 187

BT - Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012

ER -

He T, Chen S, Kim H, Tong L, Lee KW. Scheduling parallel tasks onto opportunistically available cloud resources. In Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012. 2012. p. 180-187. 6253504. (Proceedings - 2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012). https://doi.org/10.1109/CLOUD.2012.15