Abstract
The advent of multicore processors advocates for a hybrid programming model like MPI+OpenMP. Therefore, OpenMP runtimes require solid performance from a small number of threads (one MPI task per socket, OpenMP inside each socket) to a large number of threads (one MPI task per node, OpenMP inside each node). To tackle this issue, we propose a mechanism to improve performance of thread synchronization with a large spectrum of threads. It relies on a hierarchical tree traversed in a different manner according to the number of threads inside the parallel region. Our approach exposes high performance for thread activation (parallel construct) and thread synchronization (barrier construct). Several papers study hierarchical structures to launch and synchronize OpenMP threads [1, 2]. They tested tree-based approaches to distribute and synchronize threads, but they do not explore mixed hierarchical solutions.
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.
References
Nanjegowda, R., Hernandez, O., Chapman, B., Jin, H.H.: Scalability Evaluation of Barrier Algorithms for OpenMP. In: Müller, M.S., de Supinski, B.R., Chapman, B.M. (eds.) IWOMP 2009. LNCS, vol. 5568, pp. 42–52. Springer, Heidelberg (2009)
Broquedis, F., Furmento, N., Goglin, B., Wacrenier, P.A., Namyst, R.: ForestGOMP: an efficient OpenMP environment for NUMA architectures. International Journal on Parallel Programming 38(5), 418–439 (2010)
Carribault, P., Pérache, M., Jourdren, H.: Enabling Low-Overhead Hybrid MPI/OpenMP Parallelism with MPC. In: Sato, M., Hanawa, T., Müller, M.S., Chapman, B.M., de Supinski, B.R. (eds.) IWOMP 2010. LNCS, vol. 6132, pp. 1–14. Springer, Heidelberg (2010)
Bull, J.M., O’Neill, D.: A Microbenchmark Suite for OpenMP 2.0. SIGARCH Comput. Archit. News 29(5), 41–48 (2001)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Mahéo, A., Koliaï, S., Carribault, P., Pérache, M., Jalby, W. (2012). Adaptive OpenMP for Large NUMA Nodes. In: Chapman, B.M., Massaioli, F., Müller, M.S., Rorro, M. (eds) OpenMP in a Heterogeneous World. IWOMP 2012. Lecture Notes in Computer Science, vol 7312. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-30961-8_20
Download citation
DOI: https://doi.org/10.1007/978-3-642-30961-8_20
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-30960-1
Online ISBN: 978-3-642-30961-8
eBook Packages: Computer ScienceComputer Science (R0)