Abstract
We present BDDT, a task-parallel runtime system that dynamically discovers and resolves dependencies among parallel tasks. BDDT allows the programmer to specify detailed task footprints on any memory address range, multidimensional array tile or dynamic region. BDDT uses a block-based dependence analysis with arbitrary granularity. The analysis is applicable to existing C programs without having to restructure object or array allocation, and provides flexibility in array layouts and tile dimensions.
We evaluate BDDT using a representative set of benchmarks, and we compare it to SMPSs (the equivalent runtime system in StarSs) and OpenMP. BDDT performs comparable to or better than SMPSs and is able to cope with task granularity as much as one order of magnitude finer than SMPSs. Compared to OpenMP, BDDT performs up to 3.9× better for benchmarks that benefit from dynamic dependence analysis. BDDT provides additional data annotations to bypass dependence analysis. Using these annotations, BDDT outperforms OpenMP also in benchmarks where dependence analysis does not discover additional parallelism, thanks to a more efficient implementation of the runtime system.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Augonnet, C., Thibault, S., Namyst, R.: StarPU: a runtime system for scheduling tasks over accelerator-based multicore machines. Tech. Report RR-7240, INRIA (March 2010)
Ayguadé, E., Copty, N., Duran, A., Hoeflinger, J., Lin, Y., Massaioli, F., Teruel, X., Unnikrishnan, P., Zhang, G.: The Design of OpenMP Tasks. TPDS 20(3), 404–418 (2009)
Bauer, M., Clark, J., Schkufza, E., Aiken, A.: Programming the Memory Hierarchy Revisited: Supporting Irregular Parallelism in Sequoia. In: PPoPP (2011)
Best, M.J., Mottishaw, S., Mustard, C., Roth, M., Fedorova, A., Brownsword, A.: Synchronization via Scheduling: Techniques for Efficiently Managing Shared State. In: PLDI (2011)
Bienia, C., Kumar, S., Pal Singh, J., Li, K.: The PARSEC benchmark suite: Characterization and architectural implications. In: PACT (October 2008)
Bocchino, R., Adve, V.S., Dig, D., Adve, S.V., Heumann, S., Komuravelli, R., Overbey, J., Simmons, P., Sung, H., Vakilian, M.: A type and effect system for deterministic parallel Java. In: OOPSLA (2009)
Cao Minh, C., Chung, J., Kozyrakis, C., Olukotun, K.: STAMP: Stanford transactional applications for multi-processing. In: IISWC (September 2008)
Fatahalian, K., Horn, D.R., Knight, T.J., Leem, L., Houston, M., Park, J.Y., Erez, M., Ren, M., Aiken, A., Dally, W.J., Hanrahan, P.: Sequoia: Programming the Memory Hierarchy. In: SC (2006)
Herlihy, M., Moss, J.E.: Transactional memory: Architectural support for lock-free data structures. In: ISCA (1993)
Jenista, J.C., Eom, Y.H., Demsky, B.: OoOJava: Software Out-of-Order Execution. In: PPoPP (2011)
Leiserson, C.E.: The Cilk++ concurrency platform. TJS 51(3), 244–257 (2010)
Pérez, J.M., Badia, R.M., Labarta, J.: Handling Task Dependencies under Strided and Aliased References. In: ICS (2010)
Pérez, J.M., Bellens, P., Badia, R.M., Labarta, J.: CellSs: Making it Easier to Program the Cell Broadband Engine Processor. IBMRD 51(5), 593–604 (2007)
Planas, J., Badia, R.M., Ayguadé, E., Labarta, J.: Hierarchical Task-Based Programming With StarSs. IJHPCA 23(3), 284–299 (2009)
Pratikakis, P., Vandierendonck, H., Lyberis, S., Nikolopoulos, D.S.: A programming model for deterministic task parallelism. In: MSPC, pp. 7–12 (2011)
Rinard, M.C., Lam, M.S.: The Design, Implementation, and Evaluation of Jade. TOPLAS 20(3), 483–545 (1998)
Tofte, M., Talpin, J.-P.: Region-based memory management. Inf. Comput. 132(2) (1997)
Vandierendonck, H., Pratikakis, P., Nikolopoulos, D.S.: Parallel programming of general-purpose programs using task-based programming models. In: HotPar (2011)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Tzenakis, G., Papatriantafyllou, A., Vandierendonck, H., Pratikakis, P., Nikolopoulos, D.S. (2013). BDDT: Block-Level Dynamic Dependence Analysis for Task-Based Parallelism. In: Wu, C., Cohen, A. (eds) Advanced Parallel Processing Technologies. APPT 2013. Lecture Notes in Computer Science, vol 8299. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-45293-2_2
Download citation
DOI: https://doi.org/10.1007/978-3-642-45293-2_2
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-45292-5
Online ISBN: 978-3-642-45293-2
eBook Packages: Computer ScienceComputer Science (R0)