Abstract
Many computational solutions can be expressed as directed acyclic graphs (DAGs) with weighted nodes. In parallel computing, scheduling such DAGs onto manycore processors remains a fundamental challenge, since synchronization across dozens of threads and preserving precedence constraints can dramatically degrade the performance. In order to improve scheduling performance on manycore processors, we propose a hierarchical scheduling method with dynamic thread grouping, which schedules DAG structured computations at three different levels. At the top level, a supermanager separates threads into groups, each consisting of a manager thread and several worker threads. The supermanager dynamically merges and partitions the groups to adapt the scheduler to the input task dependency graphs. Through group merging and partitioning, the proposed scheduler can dynamically adjust to become a centralized scheduler, a distributed scheduler or somewhere in between, depending on the input graph. At the group level, managers collaboratively schedule tasks for their workers. At the within-group level, workers perform self-scheduling within their respective groups and execute tasks. We evaluate the proposed scheduler on the Sun UltraSPARC T2 (Niagara 2) platform that supports up to 64 hardware threads. With respect to various input task dependency graphs, the proposed scheduler exhibits superior performance when compared with other various baseline methods, including typical centralized and distributed schedulers.
This research was partially supported by the National Science Foundation under grant number CNS-0613376. NSF equipment grant CNS-0454407 is gratefully acknowledged.
Access provided by Autonomous University of Puebla. Download to read the full chapter text
Chapter PDF
Similar content being viewed by others
References
Ahmad, I., Ranka, S., Khan, S.: Using game theory for scheduling tasks on multi-core processors for simultaneous optimization of performance and energy. In: Intl. Sym. on Parallel Dist. Proc., pp. 1–6 (2008)
Kwok, Y.K., Ahmad, I.: Static scheduling algorithms for allocating directed task graphs to multiprocessors. ACM Computing Surveys 31(4), 406–471 (1999)
Zhu, W., Thulasiraman, P., Thulasiram, R.K., Gao, G.R.: Exploring financial applications on many-core-on-a-chip architecture: A first experiment. In: Frontiers of High Performance Computing and Networking, pp. 221–230 (2006)
Sheahan, D.: Developing and tuning applications on UltraSPARC T1 chip multithreading systems. Technical report (2007)
Tan, G., Sreedhar, V.C., Gao, G.R.: Analysis and performance results of computing betwenness centrality on ibm cyclops64. Journal of Supercomputing (2009)
Ahmad, I., Kwok, Y.K., Wu, M.Y.: Analysis, evaluation, and comparison of algorithms for scheduling task graphs on parallel processors. In: Proceedings of the 1996 International Symposium on Parallel Architectures, Algorithms and Networks, pp. 207–213 (1996)
Garey, M.R., Johnson, D.S.: Computers and Intractability; A Guide to the Theory of NP-Completeness. W. H. Freeman & Co., New York (1990)
Papadimitriou, C., Yannakakis, M.: Towards an architecture-independent analysis of parallel algorithms. In: Proceedings of the Twentieth Annual ACM Symposium on Theory of Computing, pp. 510–513 (1988)
Benoit, A., Hakem, M., Robert, Y.: Contention awareness and fault-tolerant scheduling for precedence constrained tasks in heterogeneous systems. Parallel Computing 35(2), 83–108 (2009)
Song, F., YarKhan, A., Dongarra, J.: Dynamic task scheduling for linear algebra algorithms on distributed-memory multicore systems. In: International Conference for Hight Performance Computing, Networking Storage and Analysis (2009)
Coffman, E.G.: Computer and Job-Shop Scheduling Theory. John Wiley and Sons, New York (1976)
Karamcheti, V., Chien, A.: A hierarchical load-balancing framework for dynamic multithreaded computations. In: Proceedings of the ACM/IEEE Conference on Supercomputing, pp. 1–17 (1998)
Zhao, H., Sakellariou, R.: Scheduling multiple DAGs onto heterogeneous systems. In: IEEE International Symposium on Parallel and Distributed Processing (IPDPS), pp. 1–12 (2006)
Blumofe, R.D., Joerg, C.F., Kuszmaul, B.C., Leiserson, C.E., Randall, K.H., Zhou, Y.: Cilk: An efficient multithreaded runtime system. Technical report, Cambridge (1996)
Intel Threading Building Blocks, http://www.threadingbuldingblocks.org/
OpenMP Application Programming Interface, http://www.openmp.org/
Charm++ programming system, http://charm.cs.uiuc.edu/research/charm/
Ohara, M., Inoue, H., Sohda, Y., Komatsu, H., Nakatani, T.: Mpi microtask for programming the cell broadband enginetm processor. IBM Systems Journal 45(1), 85–102 (2006)
Kurzak, J., Dongarra, J.: Fully dynamic scheduler for numerical computing on multicore processors. Technical report (2009)
Xia, Y., Feng, X., Prasanna, V.K.: Parallel evidence propagation on multicore processors. In: The 10th International Conference on Parallel Computing Technologies, pp. 377–391 (2009)
Bader, D.: High-performance algorithm engineering for large-scale graph problems and computational biology. In: 4th International Workshop on Efficient and Experimental Algorithms, pp. 16–21 (2005)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Xia, Y., Prasanna, V.K., Li, J. (2010). Hierarchical Scheduling of DAG Structured Computations on Manycore Processors with Dynamic Thread Grouping. In: Frachtenberg, E., Schwiegelshohn, U. (eds) Job Scheduling Strategies for Parallel Processing. JSSPP 2010. Lecture Notes in Computer Science, vol 6253. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-16505-4_9
Download citation
DOI: https://doi.org/10.1007/978-3-642-16505-4_9
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-16504-7
Online ISBN: 978-3-642-16505-4
eBook Packages: Computer ScienceComputer Science (R0)