Abstract
Maximizing system reliability and minimizing schedule length are the two major objectives in scheduling a distributed computing system. These two objectives have been considered separately by most researchers, although more realistically they should be considered simultaneously. This paper addresses the problem by taking a multi-objective approach in scheduling. A Tabu search algorithm is proposed and two lateral interference schemes are used to distribute the Pareto optimal solutions along the Pareto front uniformly. Randomly generated directed acyclic graphs and a real application task graph are used to study the performance of the proposed algorithms. Experimental results show that for this problem lateral interference has no influence on the non-dominated solution number, but does benefit the uniform distribution of non-dominated solutions, irrespective of the computation method used to determine distances between the solutions.
Similar content being viewed by others
References
Ahmad I (1995) Task assignment using a problem-space genetic algorithm. Concurr Pract Exp 7:411–428
Ben-Tal A (1980) Characterization of pareto and lexicographic optimal solutions. Multiple Criteria Dec Mak Theory Appl 177:1–11
Bozdag D, Ozguner F, Catalyurek UV (2009) Compaction of schedules and a two-stage approach for duplication-based DAG scheduling. IEEE Trans Parallel Distrib Syst 20(6):857–871
Buffet O, Cucu L, Idoumghar L, Schott R (2010) Tabu search type algorithms for the multiprocessor scheduling problem. In: 10th IASTED international conference on artificial intelligence and applications
Chitra P, Venkatesh P (2010) Multiobjective evolutionary computation algorithms for solving task scheduling problem on heterogeneous systems. Int J Knowl Based Intell Eng Syst 14(1):21–30
Chitra P, Rajaram R, Venkatesh P (2011) Application and comparison of hybrid evolutionary multiobjective optimization algorithms for solving task scheduling problem on heterogeneous systems. Appl Soft Comput 11(2):2725–2734
Coello CAC, Lamont GB, Van Veldhuisen DA (2007) Evolutionary algorithms for solving multi-objective problems. Springer, Berlin
Daoud MI, Kharma N (2008) A high performance algorithm for static task scheduling in heterogeneous distributed computing systems. J Parallel Distrib Comput 68(4):399–409
Darbha S, Agrawal DP (1998) Optimal scheduling algorithm for distributed-memory machines. IEEE Trans Parallel Distrib Syst 9(1):87–95
Dogan A, Ozguner F (2002) Matching and scheduling algorithms for minimizing execution time and failure probability of applications in heterogeneous computing. IEEE Trans Parallel Distrib Syst 13(3):308–323
da Fonseca CMM (1995) Multi-objective genetic algorithms with application to control engineering problems. PhD thesis, University of Sheffield
da Fonseca CMM, Fleming PJ (1995) An overview of evolutionary algorithms in multiobjective optimization. Evolu Comput 3(1):1–16
Erschler J, Roubellat F, Vernhes J (1976) Technical notefinding some essential characteristics of the feasible solutions for a scheduling problem. Operat Res 24(4):774–783
Fox MS (1987) Constraint-directed search: a case study of job-shop scheduling. Pitman
Garey MR, Johnson DS (1979) Computers and tntractability: a guide to the theory of NP-completeness. W. H Freeman, New York
Glover F (1997) Tabu search. Kluwer, Boston
Hou ES, Ansari N, Ren H (1994) A genetic algorithm for multiprocessor scheduling. IEEE Trans Parallel Distrib Syst 5(2):113–120
Iverson M (1999) Dynamic mapping and scheduling algorithms for a multi-user heterogeneous computing environment. PhD thesis, The Ohio State University
Jaffrs-Runser K, Gorce J, Comaniciu C (2008) A multiobjective tabu framework for the optimization and evaluation of wireless systems. Chapter 2. In: Jaziri W (ed) Tabu search. I-Tech Education and Publishing, Austria
Kartik S, Siva Ram Murthy C (1997) Task allocation algorithms for maximizing reliability of distributed computing systems. IEEE Trans Comput 46(6):719–724
KulturelKonak S, Smith AE, Norman BA (2006) Multi-objective tabu search using a multinomial probability mass function. Eur J Operat Res 169:918–931
Kwok YK, Ahmad I (1997) Efficient scheduling of arbitrary task graphs to multiprocessors using a parallel genetic algorithm. IEEE Trans Parallel Distrib Syst 47(1):58–77
Kwok YK, Ahmad I (1999) Benchmarking and comparison of the task graph scheduling algorithms. J Parallel Distrib Comput 59(3):381–422
Oh J, Wu C (2004) Genetic-algorithm-based real-time task scheduling with multiple goals. J Syst and Softw 71(3):245–258
Page AJ, Keane TM, Naughton TJ (2010) Multi-heuristic dynamic task allocation using genetic algorithms in a heterogeneous distributed system. J Parallel Distrib Comput 70(7):758–766
Qiu M, Deng J, Sha EM (2008) Failure rate minimization with multiple function unit scheduling for heterogeneous wsns. In: IEEE global telecommunications conference (GLOBECOM), pp 1–5
Sardiña IM, Boeres C, Drummond da L (2010) An efficient weighted bi-objective scheduling algorithm for heterogeneous systems. In: Euro-Par parallel processing workshops, pp 102–111
Sedaghat N, Tabatabaee-Yazdi H, Akbarzadeh T, Mohammad R (2010) Pareto front based realistic soft real-time task scheduling with multi-objective genetic algorithm in unstructured heterogeneous distributed system. In: Advances in grid and pervasive computing, pp 268–279
Shatz SM, Wang JP (1989) Models and algorithms for reliability-oriented task-allocation in redundant distributed-computer systems. IEEE Trans Reliab 38(1):16–27
Shatz SM, Wang JP, Goto M (1992) Task allocation for maximizing reliability of distributed computer systems. IEEE Trans Comput 41(9):1156–1168
Srinivas N, Deb K (1994) Muiltiobjective optimization using nondominated sorting in genetic algorithms. Evol Comput 2(3):221–248
Srinivasan S, Jha NK (1999) Safety and reliability driven task allocation in distributed systems. IEEE Trans Parallel Distrib Syst 10(3):238–251
Tan KC, Lee TH, Khor EF (2002) Evolutionary algorithms for multi-objective optimization: performance assessments and comparisons. Artif Intell Rev 17(4):251–290
Tan KC, Khor EF, Lee TH, Yang Y (2003) A tabu-based exploratory evolutionary algorithm for multiobjective optimization. Artif Intell Rev 19(3):231–260
Tang X, Li K, Li R, Veeravalli B (2010a) Reliability-aware scheduling strategy for heterogeneous distributed computing systems. J Parallel Distrib Comput 70(9):941–952
Tang X, Li K, Liao G, Li R (2010b) List scheduling with duplication for heterogeneous computing systems. J Parallel Distrib Comput 70(4):323–329
Topcuoglu H, Hariri S (2002) Performance-effective and low-complexity task scheduling for heterogeneous computing. IEEE Trans Parallel Distrib Syst 13(3):260–274
Woodside CM, Monforton GG (1993) Fast allocation of processes in distributed and parallel systems. IEEE Trans Parallel Distrib Syst 4(2):164–174
Wu M, Gajski DD (1990) Hypertool: a programming aid for message-passing systems. IEEE Trans Parallel Distrib Syst 1(3):330–343
Zhao H, Sakellariou R (2006) Scheduling multiple DAGs onto heterogeneous systems. In: International symposium on parallel and distributed processing (IPDPS), pp 14–27
Acknowledgments
Guoquan Liu acknowledges support from the start-up grant from Xi’an Jiaotong-Liverpool University. Yifeng Zeng acknowledges support from NSFC (#61375070).
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by V. Loia.
Appendix
Appendix
We provide extra results on examining the parameters in Algorithm 1.
1.1 Numbers for intensification and diversification strategies
We follow the graph setting in Table 2, and test the impact of the number for intensification and diversification strategies in Table 3.
Given the settings of (10, 300) and (40, 700) for the numbers of intensification and diversification strategies, we show the comparison of three schemes in Tables 6 and 7 respectively. The results verify better performance using either SA or SD.
1.2 Large task graph of the Gaussian elimination
We further test the Tabu search algorithm in a large real-world problem where the matrix size (\(m\)) is increased to 70 and 100. The results of comparing three schemes are shown in Tables 8 and 9 respectively. The results follow the same performance of using different schemes as that for \(m\) = 50.
Rights and permissions
About this article
Cite this article
Liu, G., Zeng, Y., Li, D. et al. Schedule length and reliability-oriented multi-objective scheduling for distributed computing. Soft Comput 19, 1727–1737 (2015). https://doi.org/10.1007/s00500-014-1360-3
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-014-1360-3