Self-Adapting Scheduling for Tasks with Dependencies in Stochastic Environments. Riakotakis, I., Ciorba, F. M., Andronikos, T., & Papakonstantinou, G. In Proceedings of the 8th IEEE International Conference on Cluster Computing (Cluster 2006) Workshops, Algorithms, Models and Tools for Parallel Computing on Heterogeneous Networks Workshop (HeteroPar 2006) , pages 1–8, September, 2006. doi abstract bibtex This paper addresses dynamic load balancing algorithms for non-dedicated heterogeneous clusters of workstations. We propose an algorithm called self-adapting scheduling (SAS), targeted at nested loops with dependencies in a stochastic environment. This means that the load entering the system, not belonging to the parallel application under execution, follows an unpredictable pattern which can be modeled by a stochastic process. SAS takes into account the history of previous timing results and the load patterns in order to make accurate load balancing predictions. We study the performance of SAS in comparison with DTSS. We established in previous work that DTSS is the most efficient self-scheduling algorithm for loops with dependencies on heterogeneous clusters. We test our algorithm under the assumption that the interarrival times and life-times of incoming jobs are exponentially distributed. The experimental results show that SAS significantly outperforms DTSS especially with rapidly varying loads
@inproceedings{riakiotakis:2006,
Abstract = {This paper addresses dynamic load balancing algorithms for non-dedicated heterogeneous clusters of workstations. We propose an algorithm called self-adapting scheduling (SAS), targeted at nested loops with dependencies in a stochastic environment. This means that the load entering the system, not belonging to the parallel application under execution, follows an unpredictable pattern which can be modeled by a stochastic process. SAS takes into account the history of previous timing results and the load patterns in order to make accurate load balancing predictions. We study the performance of SAS in comparison with DTSS. We established in previous work that DTSS is the most efficient self-scheduling algorithm for loops with dependencies on heterogeneous clusters. We test our algorithm under the assumption that the interarrival times and life-times of incoming jobs are exponentially distributed. The experimental results show that SAS significantly outperforms DTSS especially with rapidly varying loads},
Author = {Riakotakis, Ioannis and Ciorba, Florina M. and Andronikos, Theodore and Papakonstantinou, George},
Booktitle = {Proceedings of the 8th IEEE International Conference on Cluster Computing (Cluster 2006) Workshops, Algorithms, Models and Tools for Parallel Computing on Heterogeneous Networks Workshop (HeteroPar 2006) },
Date-Added = {2016-01-06 13:03:02 +0000},
Date-Modified = {2016-01-06 13:05:15 +0000},
Doi = {10.1109/CLUSTR.2006.311912},
Issn = {1552-5244},
Keywords = {2006; resource allocation;scheduling;workstation clusters;dynamic load balancing;nested loops;nondedicated heterogeneous cluster;self-adapting scheduling;stochastic environment;workstation cluster;Clustering algorithms;Heuristic algorithms;History;Life testing;Load management;Scheduling algorithm;Stochastic processes;Synthetic aperture sonar;Timing;Workstations;dynamic algorithms;load balancing;loops with dependencies;stochastic environments},
Month = {September},
Pages = {1--8},
Title = {{Self-Adapting Scheduling for Tasks with Dependencies in Stochastic Environments}},
Year = {2006},
Bdsk-Url-1 = {http://dx.doi.org/10.1109/CLUSTR.2006.311912}}
%% This BibTeX bibliography file was created using BibDesk.
%% http://bibdesk.sourceforge.net/
%% Created for Florina M. Ciorba at 2017-04-07 09:55:07 +0200
%% Saved with string encoding Unicode (UTF-8)
Downloads: 0
{"_id":{"_str":"538590720e577e1d6b002308"},"__v":57,"authorIDs":["54658f19bc7d6a460d0005a8"],"author_short":["Riakotakis, I.","Ciorba, F. M.","Andronikos, T.","Papakonstantinou, G."],"bibbaseid":"riakotakis-ciorba-andronikos-papakonstantinou-selfadaptingschedulingfortaskswithdependenciesinstochasticenvironments-2006","bibdata":{"bibtype":"inproceedings","type":"inproceedings","abstract":"This paper addresses dynamic load balancing algorithms for non-dedicated heterogeneous clusters of workstations. We propose an algorithm called self-adapting scheduling (SAS), targeted at nested loops with dependencies in a stochastic environment. This means that the load entering the system, not belonging to the parallel application under execution, follows an unpredictable pattern which can be modeled by a stochastic process. SAS takes into account the history of previous timing results and the load patterns in order to make accurate load balancing predictions. We study the performance of SAS in comparison with DTSS. We established in previous work that DTSS is the most efficient self-scheduling algorithm for loops with dependencies on heterogeneous clusters. We test our algorithm under the assumption that the interarrival times and life-times of incoming jobs are exponentially distributed. The experimental results show that SAS significantly outperforms DTSS especially with rapidly varying loads","author":[{"propositions":[],"lastnames":["Riakotakis"],"firstnames":["Ioannis"],"suffixes":[]},{"propositions":[],"lastnames":["Ciorba"],"firstnames":["Florina","M."],"suffixes":[]},{"propositions":[],"lastnames":["Andronikos"],"firstnames":["Theodore"],"suffixes":[]},{"propositions":[],"lastnames":["Papakonstantinou"],"firstnames":["George"],"suffixes":[]}],"booktitle":"Proceedings of the 8th IEEE International Conference on Cluster Computing (Cluster 2006) Workshops, Algorithms, Models and Tools for Parallel Computing on Heterogeneous Networks Workshop (HeteroPar 2006) ","date-added":"2016-01-06 13:03:02 +0000","date-modified":"2016-01-06 13:05:15 +0000","doi":"10.1109/CLUSTR.2006.311912","issn":"1552-5244","keywords":"2006; resource allocation;scheduling;workstation clusters;dynamic load balancing;nested loops;nondedicated heterogeneous cluster;self-adapting scheduling;stochastic environment;workstation cluster;Clustering algorithms;Heuristic algorithms;History;Life testing;Load management;Scheduling algorithm;Stochastic processes;Synthetic aperture sonar;Timing;Workstations;dynamic algorithms;load balancing;loops with dependencies;stochastic environments","month":"September","pages":"1–8","title":"Self-Adapting Scheduling for Tasks with Dependencies in Stochastic Environments","year":"2006","bdsk-url-1":"http://dx.doi.org/10.1109/CLUSTR.2006.311912","bibtex":"@inproceedings{riakiotakis:2006,\n\tAbstract = {This paper addresses dynamic load balancing algorithms for non-dedicated heterogeneous clusters of workstations. We propose an algorithm called self-adapting scheduling (SAS), targeted at nested loops with dependencies in a stochastic environment. This means that the load entering the system, not belonging to the parallel application under execution, follows an unpredictable pattern which can be modeled by a stochastic process. SAS takes into account the history of previous timing results and the load patterns in order to make accurate load balancing predictions. We study the performance of SAS in comparison with DTSS. We established in previous work that DTSS is the most efficient self-scheduling algorithm for loops with dependencies on heterogeneous clusters. We test our algorithm under the assumption that the interarrival times and life-times of incoming jobs are exponentially distributed. The experimental results show that SAS significantly outperforms DTSS especially with rapidly varying loads},\n\tAuthor = {Riakotakis, Ioannis and Ciorba, Florina M. and Andronikos, Theodore and Papakonstantinou, George},\n\tBooktitle = {Proceedings of the 8th IEEE International Conference on Cluster Computing (Cluster 2006) Workshops, Algorithms, Models and Tools for Parallel Computing on Heterogeneous Networks Workshop (HeteroPar 2006) },\n\tDate-Added = {2016-01-06 13:03:02 +0000},\n\tDate-Modified = {2016-01-06 13:05:15 +0000},\n\tDoi = {10.1109/CLUSTR.2006.311912},\n\tIssn = {1552-5244},\n\tKeywords = {2006; resource allocation;scheduling;workstation clusters;dynamic load balancing;nested loops;nondedicated heterogeneous cluster;self-adapting scheduling;stochastic environment;workstation cluster;Clustering algorithms;Heuristic algorithms;History;Life testing;Load management;Scheduling algorithm;Stochastic processes;Synthetic aperture sonar;Timing;Workstations;dynamic algorithms;load balancing;loops with dependencies;stochastic environments},\n\tMonth = {September},\n\tPages = {1--8},\n\tTitle = {{Self-Adapting Scheduling for Tasks with Dependencies in Stochastic Environments}},\n\tYear = {2006},\n\tBdsk-Url-1 = {http://dx.doi.org/10.1109/CLUSTR.2006.311912}}\n\n%% This BibTeX bibliography file was created using BibDesk.\n%% http://bibdesk.sourceforge.net/\n\n%% Created for Florina M. Ciorba at 2017-04-07 09:55:07 +0200 \n\n\n%% Saved with string encoding Unicode (UTF-8) \n\n\n\n","author_short":["Riakotakis, I.","Ciorba, F. M.","Andronikos, T.","Papakonstantinou, G."],"bibbaseid":"riakotakis-ciorba-andronikos-papakonstantinou-selfadaptingschedulingfortaskswithdependenciesinstochasticenvironments-2006","role":"author","urls":{},"keyword":["2006; resource allocation;scheduling;workstation clusters;dynamic load balancing;nested loops;nondedicated heterogeneous cluster;self-adapting scheduling;stochastic environment;workstation cluster;Clustering algorithms;Heuristic algorithms;History;Life testing;Load management;Scheduling algorithm;Stochastic processes;Synthetic aperture sonar;Timing;Workstations;dynamic algorithms;load balancing;loops with dependencies;stochastic environments"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://bibbase.org/f/5soafvAi5dxziNqBd/conferences.bib","downloads":0,"keywords":["2006; resource allocation;scheduling;workstation clusters;dynamic load balancing;nested loops;nondedicated heterogeneous cluster;self-adapting scheduling;stochastic environment;workstation cluster;clustering algorithms;heuristic algorithms;history;life testing;load management;scheduling algorithm;stochastic processes;synthetic aperture sonar;timing;workstations;dynamic algorithms;load balancing;loops with dependencies;stochastic environments"],"search_terms":["self","adapting","scheduling","tasks","dependencies","stochastic","environments","riakotakis","ciorba","andronikos","papakonstantinou"],"title":"Self-Adapting Scheduling for Tasks with Dependencies in Stochastic Environments","year":2006,"dataSources":["gXaEZgfAnH6LvDp7P","yquS8QakKCWDYDAk9","aMnnnpEwMvMZwRFdC","we4PjHBkvY3ihPaQy","yezXfWaY6nTQG2nAv","pWKBSwppQZXZa3qnh"]}