Grzegorz Malewicz
ABSTRACT
This paper introduces a parallel scheduling problem where a directed acyclic graph modeling t tasks and their dependencies needs to be executed on n unreliable workers. Worker i executes task j correctly with probability pi,j. The goal is to find a regimen Ε, that dictates how workers get assigned to tasks (possibly in parallel and redundantly) throughout execution, so as to minimize expected completion time. This fundamental parallel scheduling problem arises in grid computing and project management fields, and has several practical applications.We show a polynomial time algorithm for the problem restricted to the case when dag width is at most a constant and the number of workers is also at most a constant. These two restrictions may appear to be too severe. However, they are fundamentally required. Specifically, we demonstrate that the problem is NP-hard with constant number of workers when dag width can grow, and is also NP-hard with constant dag width when the number of workers can grow. When both dag width and the number of workers are unconstrained, then the problem is inapproximable within factor less than 5/4, unless P=NP.
- Annis, J., Zhao, Y., Voeckler, J., Wilde, M., Kent, S., Foster, I.: Applying Chimera Virtual Data Concepts to Cluster Finding in the Sloan Sky Survey. 15th Conference on High Performance Networking and Computing (SC) (2002) 56. Google Scholar
Digital Library
- Cook, S.A.: An observation on time-storage tradeoff. Journal of Computer and System Sciences, Vol. 9 (1974) 308--316.Google ScholarDigital Library
- Cormen, T.H., Leiserson, C.E., Rivest, R.L., Stein, C.: Introduction to Algorithms (2nd Edition). MIT Press (2001). Google ScholarDigital Library
- Crescenzi, P., Kann, V. (eds.): A compendium of NP optimization problems. http://www.nada.kth.se/~viggo/wwwcompendium/node173.htmlGoogle Scholar
- Dilworth, R.P.: A decomposition theorem for partially ordered sets. Annals of Mathematics, Vol. 51 (1950) 161--166.Google ScholarCross Ref
- Fernandez, A., Armacost, R., Pet-Edwards, J.: Understanding Simulation Solutions to Resource constrained Project Scheduling Problems with Stochastic task Durations. Engineering Management Journal, Vol. 10(4) (1998) 5--13.Google ScholarCross Ref
- Fernandez, A., Armacost, R.L., Pet-Edwards, J.: A Model for the Resource Constrained Project Scheduling Problem with Stochastic Task Durations. 7th Industrial Engineering Research Conference Proceedings (1998).Google Scholar
- Foster, I., Kesselman, C. {eds.}: The Grid: Blueprint for a New Computing Infrastructure, 2nd ed. Morgan-Kaufmann, San Francisco, CA (2004). Google ScholarDigital Library
- Gao, L., Malewicz, G.: Internet computing of tasks with dependencies using unreliable workers. 8th International Conference on Principles of Distributed Systems (OPODIS) (2004) 315--325. Google ScholarDigital Library
- Garey, M.R., Johnson, D.S.: Computers and Intractability. Freeman, New York (1979). Google ScholarDigital Library
- Goel, A., Indyk, P.: Stochastic load balancing and related problems. 40th Annual Symposium on Foundations of Computer Science (FOCS) (1999) 579--586. Google ScholarDigital Library
- Hillier, F.S., Lieberman, G.J.: Introduction to Operations Research, 8th ed. McGraw-Hill (2004). Google ScholarDigital Library
- Herroelen, W., Leus, R.: Project scheduling under uncertainty: Survey and research potentials. European Journal of Operational Research, Vol. 165(2) (2005) 289--306.Google ScholarCross Ref
- Hong, J.-W., Kung, H.T.: I/O complexity: the red-blue pebble game. 13th ACM Symposium on Theory of Computing (STOC) (1981) 326--333. Google ScholarDigital Library
- Hopcroft, J.E., Paul, W., Valiant, L.G.: On time versus space. Journal of the ACM, Vol. 24 (1977) 332--337. Google ScholarDigital Library
- Ireland, K., Rosen, M.: A Classical Introduction to Modern Number Theory, 2nd ed. Springer-Verlag (1990).Google Scholar
- Johnson, D.S.: The NP-completeness column: An ongoing guide. Journal of Algorithms, Vol. 8(3) (1987) 438--448. Google ScholarDigital Library
- Kleinberg, J., Rabani, Y., Tardos, E.: Allocating Bandwidth for Bursty Connections. SIAM Journal on Computing, Vol. 30(1) (2000) 191--217. Google ScholarDigital Library
- Malewicz, G.: Implementation and Experiments with an Algorithm for Parallel Scheduling of Complex Dags under Uncertainty. (2005) submitted for publication.Google Scholar
- Malewicz, G., Rosenberg, A.L.: On batch-scheduling dags for Internet-based computing. 11th European Conference on Parallel Processing (Euro-Par) (2005) to appear. Google ScholarDigital Library
- Malewicz, G., Rosenberg, A., Yurkewych, M.: Toward a Scheduling Theory for Internet-Based Computing. Submitted for publication. Preliminary version: 19th IEEE International Parallel & Distributed Processing Symposium (IPDPS) (2005) 66. Google ScholarDigital Library
- Microsoft Project 2003 http://www.microsoft.com/office/project/default.aspGoogle Scholar
- Mori, M., Tseng, C.: A Resource Constrained Project Scheduling Problem with Reattempt at Failure: A Heuristic Approach. Journal of the Operations Research Society of Japan, Vol. 40(1) (1997) 33--44.Google ScholarCross Ref
- Narasimhan, M., Ramanujam, J.: A fast approach to computing exact solutions to the resource-constrained scheduling problem. ACM Transactions on Design Automation of Electronic Systems, Vol. 6(4) (2001) 490--500. Google ScholarDigital Library
- Özdamar, L., Ulusoy, G.: A survey on the resource-constrained project scheduling problem. IIE Transactions, Vol. 27 (1995) 574--586.Google ScholarCross Ref
- Paterson, M.S., Hewitt, C.E.: Comparative schematology. Project MAC Conference on Concurrent Systems and Parallel Computation, ACM Press (1970) 119--127.Google Scholar
- Peeters, R.: The maximum edge biclique problem is NP-complete. Discrete Applied Mathematics, Vol. 131(3) (2003) 651--654. Google ScholarDigital Library
- Rosenberg, A.L.: On scheduling mesh-structured computations for Internet-based computing. IEEE Transactions on Computers, Vol. 53 (2004) 1176--1186. Google ScholarDigital Library
- Rosenberg, A.L., Sudborough, I.H.: Bandwidth and pebbling. Computing Vol. 31 (1983) 115--139.Google ScholarCross Ref
- Rosenberg, A.L., Yurkewych, M.: Guidelines for scheduling some common computation-dags for Internet-based computing. IEEE Transactions on Computers, Vol. 54(4) (2005) 428--438. Google ScholarDigital Library
- Sarmenta, L.F.G.: Sabotage-tolerance mechanisms for volunteer computing systems. Future Generation Computer Systems, Vol. 18(4) (2002) 561--572. Google ScholarDigital Library
- Skutella, M., Uetz, M.: Scheduling precedence-constrained jobs with stochastic processing times on parallel machines. 12th ACM-SIAM Symposium on Discrete Algorithms (SODA) (2001) 589--590. Google ScholarDigital Library
- Thain, D., Tannenbaum, T., Livny, M.: Distributed Computing in Practice: The Condor Experience. Concurrency and Computation: Practice and Experience (2005) to appear. Google ScholarDigital Library
- Tseng, C.C., Mori, M., Yajima, Y.: A project scheduling model considering the success probability. Proc. of the Association of Asian Pacific Operational Research Societies (APROS) (1994) 399--406.Google Scholar
- Turnquist, M.A., Nozick, L.K.: Allocating Time and Resources in Project Management Under Uncertainty. 36th Annual Hawaii International Conference on System Sciences (HICSS) (2003) 250c Google ScholarDigital Library
Index Terms
- Parallel scheduling of complex dags under uncertainty
Recommendations
Implementation and experiments with an algorithm for parallel scheduling of complex dags under uncertainty
Proceedings of the Meeting on Algorithm Engineering & ExpermimentsOur earlier paper introduced a parallel scheduling problem where a directed acyclic graph modeling t tasks and their dependencies needs to be executed on n unreliable workers. Worker i executes task j correctly with probability pi,j. The goal is to find ...
Toward a Theory for Scheduling Dags in Internet-Based Computing
Conceptual and algorithmic tools are developed as a foundation for a theory of scheduling complex computation-dags for Internet-based computing. The goal of the schedules produced is to render tasks eligible for allocation to remote clients (hence, for ...
Single machine scheduling problems with deteriorating jobs
This paper considers the single machine scheduling problems with deteriorating jobs, i.e. jobs whose processing times are a decreasing linear function of their starting time. It is assumed that jobs have the different basic processing time and same ...
Comments