# Scheduling Stochastic Jobs on Parallel Machines to Minimize Makespan or Flowtime

## Abstract

A number of identical machines operating in parallel are to be used to complete the processing of a collection of jobs so as to minimize the jobs’ makespan or flowtime. The amount of processing required to complete the jobs have known probability distributions. It has been established by several researchers that when the required amounts of processing are all distributed as exponential random variables, then the strategy (LEPT) of always processing jobs with the longest expected processing times minimizes the expected value of the makespan, and the strategy (SEPT) of always processing jobs with the shortest expected processing times minimizes the expected value of the flowtime. We prove these results and describe a more general instance in which they are also true: when the jobs have received differing amounts of processing prior to the start, their total processing requirements are identically distributed, and the common distribution of total processing requirements has a monotone hazard rate. Under the stronger assumption that the distribution of the total processing requirements has a density whose logarithm is concave or convex, LEPT and SEPT minimize the makespan and flowtime in distribution.

## Keywords

Optimal Policy Parallel Machine Processing Requirement Identical Machine Exponential Random Variable## Preview

Unable to display preview. Download preview PDF.

## References

- [1]Bruno, J. (1976), “Sequencing Tasks with Exponential Service Times on Parallel Machines,” Technical report, Department of Computer Science, Pennsylvania State University.Google Scholar
- [2]Bruno, J. and Downey, P. (1977), “Sequencing Tasks with Exponential Service Times on Parallel Machines,” Technical report, Department of Computer Sciences, University of California at Santa Barbara.Google Scholar
- [3]Bruno, J., Downey, P. and Frederickson, G. N. (1981), “Sequencing Tasks with Exponential Service Times to Minimize the Expected Flowtime or Makespan,” J. Assoc. Comput. Mach., 28, 100–113.CrossRefGoogle Scholar
- [4]Conway, R. W., Maxwell, W. L. and Miller, L. W. (1967), The Theory of Scheduling, Addison-Wesley, Reading, Massachusetts.Google Scholar
- [5]Cox, D. R. (1959), “A Renewal Problem with Bulk Ordering of Components,” J. Roy. Statist. Soc. Ser. B, 21, 180–189.Google Scholar
- [6]Glazebrook, K. D. (1976), “Stochastic Scheduling,” Ph. D. Thesis, University of Cambridge.Google Scholar
- [7]Glazebrook, K. D. (1979), “Scheduling Tasks with Exponential Service Times on Parallel Processors,” J. Appl. Probab., 16, 685–689.CrossRefGoogle Scholar
- [8]Karlin, S. (1968), Total Positivity, Vol. I., Stanford University Press, Stanford.Google Scholar
- [9]McNaughton, R. (1959), “Scheduling with Deadline and Loss Functions,” Management Sci., 6, 1–12.CrossRefGoogle Scholar
- [10]Nash, P. (1973), “Optimal Allocation of Resources to Research Projects,” Ph. D. Thesis, University of Cambridge.Google Scholar
- [11]Pinedo, M. and Weiss, G. (1979), “Scheduling Stochastic Tasks on Two Parallel Processors,” Naval Res. Logist. Quart., 27, 528–536.Google Scholar
- [12]Schrage, L. E. (1968), “A Proof of the Shortest Remaining Process Time Discipline,” Oper. Res., 16, 687–689.CrossRefGoogle Scholar
- [13]Van der Heyden, J. (1981), “Scheduling Jobs with Exponential Processing and Arrival Times on Identical Processors so as to Minimize the Expected Makespan,” Math. Oper. Res., 6, 305–312.CrossRefGoogle Scholar
- [14]Weber, R. R. and Nash, P. (1979), “An Optimal Strategy in Multi-Server Stochastic Scheduling,” J. Roy. Statist. Soc. Ser. B, 40, 323–328.Google Scholar
- [15]Weber, R. R. (1980), “Optimal Organization of Multi-Server Systems,” Ph. D. Thesis, University of Cambridge.Google Scholar
- [16]Weber, R. R. (1982), “Scheduling Jobs with Stochastic Processing Requirements on Parallel Machines to Minimize Makespan and Flowtime,” J. Appl. Probab., 19, (to appear).Google Scholar
- [17]Weiss, G. and Pinedo, M. (1979), “Scheduling Tasks with Exponential Service Times on Non-Identical Processors to Minimize Various Cost Functions, J. Appl. Probab., 17, 187–202.CrossRefGoogle Scholar

## References

- [1]Bruno, J., P. Downey and G. N. Frederickson (1981), “Sequencing Tasks with Exponential Service Times to Minimize the Expected Flow Time or Makespan,” J. Assoc. Comput. Mach., 28, pp. 100–113.CrossRefGoogle Scholar
- [2]Chandy, K. M. and P. F. Reynolds (1975), “Scheduling Partially Ordered Tasks with Probabilistic Execution Times,” Proceedings of the 5th Symposium on Operating Systems Principles, ACM, 169-177.Google Scholar
- [3]Coffman, E. G., Jr. (ed.) (1976), Computer and Job-Shop Scheduling Theory, John Wiley and Sons, New York.Google Scholar
- [4]Conway, R. W., W. L. Maxwell and L. W. Miller (1967), Theory of Scheduling, Addison-Wesley Publishing Co., Reading, MA.Google Scholar
- [5]Garey, M. R. and D. S. Johnson (1979), Computers and Intractability: a Guide to the Theory of NP-Completeness, W. H. Freeman and Co., San Francisco, CA.Google Scholar
- [6]Glazebrook, K. D. (1976), “Stochastic Scheduling,” Ph.D. Thesis, Cambridge University.Google Scholar
- [7]Hu, T. C. (1961), “Parallel Sequencing and Assembly Line Problems,” Oper. Res., 9, pp. 841–848.CrossRefGoogle Scholar
- [8]Karlin, S. (1968), “Total Positivity, Vol. 1,” Stanford University Press, Stanford, CA.Google Scholar
- [9]Kaufmann, A., D. Grouchko and R. Crunon (1977), Mathematical Models For the Study of the Reliability of Systems, Academic Press, New York.Google Scholar
- [10]McNaughton, R. (1959), “Scheduling with Deadlines and Loss Functions, Management Sci., 16, pp. 1–12.CrossRefGoogle Scholar
- [11]Nash, P. (1973), “Optimal Allocation of Resources to Research Projects,” Ph.D. Thesis, Cambridge University.Google Scholar
- [12]Pinedo, M. and G. Weiss (1979), “Scheduling Stochastic Tasks on Two Parallel Processors,” Naval Res. Logist. Quart., 26, pp. 527–535.CrossRefGoogle Scholar
- [13]Weber, R. (1981), “Scheduling Jobs with Stochastic Processing Requirements on Parallel Machines to Minimize Makespan or Flowtime,” J. Appl. Probab., 18, to appear.Google Scholar
- [14]Weiss, G. and M. Pinedo (1980), “Scheduling Tasks with Exponential Service Times on Non-Identical Processors to Minimize Various Cost Functions, J. Appl. Probab., 17, pp. 187–202.CrossRefGoogle Scholar