Leveraging shared caches for parallel temporal blocking of stencil codes on multicore processors and clusters

Parallel Processing Letters

Volumn 20, Issue 4, 2010, Pages 359-376

  Wittmann, Markus ^a   Hager, Georg ^a   Treibig, Jan ^a   Wellein, Gerhard ^a  

^a UNIVERSITY OF ERLANGEN NUREMBERG   (Germany)

Author keywords

multi halo exchange; multicore; shared caches; stencil algorithm; temporal blocking

Indexed keywords

MULTI CORE; MULTI-HALO EXCHANGE; SHARED CACHE; STENCIL ALGORITHM; TEMPORAL BLOCKING;

ALGORITHMS; BANDWIDTH;

CACHE MEMORY;

EID: 78650871519     PISSN: 01296264     EISSN: None     Source Type: Journal    
DOI: 10.1142/S0129626410000296     Document Type: Conference Paper

References (14)

1. B. Bergen, F. Hülsemann, U. Rüde: Is 1.7×1010 Unknowns the Largest Finite El- ement System that Can Be Solved Today? In: ACM/IEEE (Ed.): Proceedings of the ACM/IEEE SC 2005 Conference (Supercomputing Conference '05, Seattle, Nov 12-18, 2005).
2. K. Datta et al.: Stencil Computation Optimization and Auto-tuning on State-of-the- art Multicore Architectures. Proc. SC2008. DOI: 10.1145/1413370.1413375
3. K. Datta et al.: Optimization and performance modeling of stencil computa- tions on modern microprocessors. SIAM Review 51(1), (2009) 129-159. DOI: 10.1137/070693199
4. G.Wellein, G. Hager, T. Zeiser, M.Wittmann, H. Fehske: Efficient temporal blocking for stencil computations by multicore-aware wavefront parallelization. Proc. COMP- SAC 2009. DOI: 10.1109/COMPSAC.2009.82
5. J. Treibig, G. Hager: Introducing a Performance Model for Bandwidth Limited Loop Kernels. Workshop on Memory Issues on Multi- and Manycore Platforms, Proc. PPAM 2009. http://arxiv.org/abs/0905.0792
6. D. Wonnacott: Using Time Skewing to Eliminate Idle Time due to Memory Bandwidth and Network Limitations. Proc. IPDPS 2000. DOI: 10.1109/IPDPS.2000. 845979
7. G. Jin, J. Mellor-Crummey, R. Fowler: Increasing temporal locality with skewing and recursive blocking. Proc. SC2001. DOI: 10.1145/582034.582077
8. M. Kowarschik: Data Locality Optimizations for Iterative Numerical Algorithms and Cellular Automata on Hierarchical Memory Architectures. Dissertation, SCS Publishing House, Erlangen (2004). http://www10.informatik. uni-erlangen.de/Publications/Dissertations/Diss-Kowarschik-2004.pdf
9. M. Frigo, V. Strumpen: Cache oblivious stencil computations. Proc. ICS 2005. DOI: 10.1145/1088149.1088197 (Pubitemid 43251341)
10. S. Kamil, K. Datta, S. Williams, L. Oliker, J. Shalf, K. Yelick: Implicit and explicit optimizations for stencil computations. Proc. MSPC 2006. DOI: 10.1145/1178597.1178605 (Pubitemid 47177302)
11. T. Zeiser, G. Wellein, A. Nitsure, K. Iglberger, U. Rüde, G. Hager: Introducing a parallel cache oblivious blocking approach for the lattice Boltzmann method. Progress in CFD, vol. 8, No. 1-4, pp. 179-188, 2008.
12. J. Treibig, G. Hager, G. Wellein: Multi-core architectures: Complexities of performance prediction and the impact of cache topology. Under review. http://arxiv. org/abs/0910.4865
13. C. Ding, Y. He: A ghost cell expansion method for reducing communications in solving PDE problems. Proc. SC2001. DOI: 10.1145/582034.582084
14. M. Wittmann: Potentials of temporal blocking for stencil-based computations on multi-core systems. Master's Thesis, University of Applied Sciences Nuremberg, March 2009. http://www.hpc.rrze.uni-erlangen.de/Projekte/ stencil.shtml