Design and Management of Complex Technical Processes and Systems by Means of Computational Intelligence Methods Scheduling Algorithm Development Based on Complex Owner Defined Objectives Scheduling Algorithm Development Based on Complex Owner Defined Objectives

This paper presents a methodology to automatically generate an online scheduling algorithm for a complex objective defined by an owner of parallel identical machines. The objective is based on a combination of simple conventional objectives. Specifically, we address a non-preemptive online scheduling problem on parallel identical machines for a small user community with independent jobs. As the scheduling problem is evaluated using simple conventional objectives for different users that are often contradicting a multiobjective approach is needed. First, Evolutionary Algorithms are applied to create an approximation of the Pareto front in a 7-dimensional space of possible schedules. This enables the machine owner to define a complex objective according to his preferences. For the actual scheduling, a simple Greedy algorithm is proposed whose parameters are determined by an Evolutionary Algorithm such that the selected objective is observed.

[1]  Uwe Schwiegelshohn,et al.  Improving First-Come-First-Serve Job Scheduling by Gang Scheduling , 1998, JSSPP.

[2]  Uwe Schwiegelshohn,et al.  On an on-line scheduling problem for parallel jobs , 2002, Inf. Process. Lett..

[3]  Ramin Yahyapour,et al.  Economic Scheduling in Grid Computing , 2002, JSSPP.

[4]  Dror G. Feitelson,et al.  Memory Usage in the LANL CM-5 Workload , 1997, JSSPP.

[5]  Uwe Schwiegelshohn,et al.  Theory and Practice in Parallel Job Scheduling , 1997, JSSPP.

[6]  Michael Pinedo,et al.  Scheduling: Theory, Algorithms, and Systems , 1994 .

[7]  Steven Hotovy,et al.  Workload Evolution on the Cornell Theory Center IBM SP2 , 1996, JSSPP.

[8]  Bill Nitzberg,et al.  A comparison of workload traces from two production parallel machines , 1996, Proceedings of 6th Symposium on the Frontiers of Massively Parallel Computation (Frontiers '96).

[9]  Cynthia Bailey Lee,et al.  Are User Runtime Estimates Inherently Inaccurate? , 2004, JSSPP.

[10]  Susanne Albers,et al.  Online algorithms: a survey , 2003, Math. Program..

[11]  Dror G. Feitelson,et al.  Job Characteristics of a Production Parallel Scientivic Workload on the NASA Ames iPSC/860 , 1995, JSSPP.

[12]  Uwe Schwiegelshohn.  Preemptive Weighted Completion Time Scheduling of Parallel Jobs , 1996, ESA.

[13]  Dror G. Feitelson,et al.  Utilization, Predictability, Workloads, and User Runtime Estimates in Scheduling the IBM SP2 with Backfilling , 2001, IEEE Trans. Parallel Distributed Syst..

[14]  Robert McNaughton,et al.  Scheduling with Deadlines and Loss Functions , 1959 .

[15]  Robert E. Tarjan,et al.  Self-adjusting binary search trees , 1985, JACM.

[16]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[17]  Dror G. Feitelson,et al.  Metric and workload effects on computer systems evaluation , 2003, Computer.

[18]  Ramin Yahyapour,et al.  Scaling of Workload Traces , 2003, JSSPP.

[19]  Partha Dasgupta,et al.  Preemptive Scheduling for Distributed Systems , 1998 .

[20]  Hans-Georg Beyer,et al.  The Theory of Evolution Strategies , 2001, Natural Computing Series.

[21]  Dror G. Feitelson,et al.  Utilization and Predictability in Scheduling the IBM SP2 with Backfilling , 1998, Proceedings of the First Merged International Parallel Processing Symposium and Symposium on Parallel and Distributed Processing.

[22]  Ramin Yahyapour.  Grid Resource Management-State of the Art and Future Trends, chapter Applying Economic Scheduling Me , 2003 .

[23]  Uwe Schwiegelshohn,et al.  Fairness in parallel job scheduling , 2000 .

[24]  Zbigniew Michalewicz,et al.  Handbook of Evolutionary Computation , 1997 .

[25]  David A. Lifka,et al.  The ANL/IBM SP Scheduling System , 1995, JSSPP.