Job Scheduling Strategies for Parallel Processing

Duplication based heuristics have been widely utilized for scheduling communication intensive, precedence constrained tasks on multiple processors. Duplicating the predecessor of a task on the processor to which the task is assigned can result in the minimization of the communication cost. This helps in reducing the schedule length. However, this reduction comes at the cost of extra computing power required to duplicate the tasks. We have tried to address this trade-off in this paper. We propose “controlled” duplication algorithms for scheduling real-time periodic tasks with end-to-end deadlines on heterogeneous multiprocessors. We observe that whether to duplicate tasks or not is decided by the task deadlines. In the case that the deadline can be met without duplication, more schedule holes are created. These holes can be used to schedule other tasks. Simulations show that the proposed algorithms efficiently utilize the holes and improve the success ratio by 15%–50% versus comparable algorithms.

[1]  A. Feinstein,et al.  High agreement but low kappa: I. The problems of two paradoxes. , 1990, Journal of clinical epidemiology.

[2]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[3]  Pascal Bouvry,et al.  Management of an academic HPC cluster: The UL experience , 2014, 2014 International Conference on High Performance Computing & Simulation (HPCS).

[4]  Arie Ben-David,et al.  Comparison of classification accuracy using Cohen's Weighted Kappa , 2008, Expert Syst. Appl..

[5]  Radu Prodan,et al.  A Hybrid Intelligent Method for Performance Modeling and Prediction of Workflow Activities in Grids , 2009, 2009 9th IEEE/ACM International Symposium on Cluster Computing and the Grid.

[6]  Jean-Marc Vincent,et al.  Analysis of the Jobs Resource Utilization on a Production System , 2013, JSSPP.

[7]  Dániel Szöllősi,et al.  Comparison of six multiclass classifiers by the use of different classification performance indicators , 2012 .

[8]  Richard Gibbons,et al.  A Historical Application Profiler for Use by Parallel Schedulers , 1997, JSSPP.

[9]  Dan Tsafrir,et al.  Backfilling Using System-Generated Predictions Rather than User Runtime Estimates , 2007, IEEE Transactions on Parallel and Distributed Systems.

[10]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[11]  A. Snavely,et al.  What ’ s working in HPC : Investigating HPC User Behavior and Productivity , 2006 .

[12]  Andrew P. Bradley,et al.  The use of the area under the ROC curve in the evaluation of machine learning algorithms , 1997, Pattern Recognit..

[13]  David H. Bailey,et al.  The Nas Parallel Benchmarks , 1991, Int. J. High Perform. Comput. Appl..

[14]  Dror G. Feitelson,et al.  The workload on parallel supercomputers: modeling the characteristics of rigid jobs , 2003, J. Parallel Distributed Comput..

[15]  Satoshi Matsuoka,et al.  Overview of a performance evaluation system for global computing scheduling algorithms , 1999, Proceedings. The Eighth International Symposium on High Performance Distributed Computing (Cat. No.99TH8469).

[16]  David J. Hand,et al.  A Simple Generalisation of the Area Under the ROC Curve for Multiple Class Classification Problems , 2001, Machine Learning.

[17]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[18]  Dror G. Feitelson,et al.  Improved Utilization and Responsiveness with Gang Scheduling , 1997, JSSPP.

[19]  Dan Tsafrir,et al.  Experience with using the Parallel Workloads Archive , 2014, J. Parallel Distributed Comput..

[20]  John S. Uebersax,et al.  A Generalized Kappa Coefficient , 1982 .

[21]  José A. B. Fortes,et al.  On the Use of Machine Learning to Predict the Time and Resources Consumed by Applications , 2010, 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing.

[22]  S. Sathiya Keerthi,et al.  Evaluation of simple performance measures for tuning SVM hyperparameters , 2003, Neurocomputing.

[23]  Dror G. Feitelson,et al.  Workload Modeling for Performance Evaluation , 2002, Performance.

[24]  Warren Smith,et al.  Predicting Application Run Times Using Historical Information , 1998, JSSPP.

[25]  Tom Fawcett,et al.  Analysis and Visualization of Classifier Performance: Comparison under Imprecise Class and Cost Distributions , 1997, KDD.

[26]  Georges Da Costa,et al.  2005 IEEE International Symposium on Cluster Computing and the Grid , 2005, CCGRID.

[27]  Junwei Cao,et al.  Queue scheduling and advance reservations with COSY , 2004, 18th International Parallel and Distributed Processing Symposium, 2004. Proceedings..

[28]  H. Casanova,et al.  The MicroGrid : Using Emulation to Predict Application Performance in Diverse Grid Network Environments , 2004 .

[29]  Isabelle Guyon,et al.  A Scaling Law for the Validation-Set Training-Set Size Ratio , 1997 .