Scalability and locality of extrapolation methods on large parallel systems

Concurrency and Computation: Practice and Experience

Volumn 23, Issue 15, 2011, Pages 1789-1815

  Korch, Matthias ^a   Rauber, Thomas ^a   Scholtes, Carsten ^a  

^a UNIVERSITY OF BAYREUTH   (Germany)

Author keywords

distributed memory; locality; NUMA; ordinary differential equations; parallel extrapolation methods; scalability; shared memory

Indexed keywords

CACHE MEMORY; EXTRAPOLATION; MEMORY ARCHITECTURE; SCALABILITY; SUPERCOMPUTERS;

DISTRIBUTED MEMORY; EXTRAPOLATION METHODS; LOCALITY; NUMA; SHARED MEMORY;

ORDINARY DIFFERENTIAL EQUATIONS;

EID: 80052925866     PISSN: 15320626     EISSN: 15320634     Source Type: Journal    
DOI: 10.1002/cpe.1765     Document Type: Article

References (25)

1. Deuflhard P,. Recent progress in extrapolation methods for ordinary differential equations. SIAM Review 1985; 27: 505-535. (Pubitemid 15607687)
2. Ehrig R, Nowak U, Deuflhard P,. Massively parallel linearly-implicit extrapolation algorithms as a powerful tool in process simulation. Parallel Computing: Fundamentals, Applications and New Directions. Elsevier: Amsterdam, 1998; 517-524.
3. Hairer E, Nørsett SP, Wanner G,. Solving Ordinary Differential Equations I: Nonstiff Problems (2nd rev. edn). Springer: Berlin, 2000.
4. Kappeller M, Kiehl M, Perzl M, Lenke M,. Optimized extrapolation methods for parallel solution of IVPs on different computer architectures. Applied Mathematics and Computation 1996; 77 (2-3): 301-315. (Pubitemid 126667288)
5. Lustman L, Neta B, Gragg W,. Solution of ordinary differential initial value problems on an Intel Hypercube. Computer and Mathematics with Applications 1992; 23 (10): 65-72.
6. Rauber T, Rnger G,. Load balancing schemes for extrapolation methods. Concurrency: Practice and Experience 1997; 9 (3): 181-202. (Pubitemid 127679697)
7. van der Houwen PJ, Sommeijer BP,. Parallel ODE solvers. Proceedings of the ACM International Conference on Supercomputing, Amsterdam, The Netherlands. ACM: New York, NY, U.S.A., 1990; 71-81. (Pubitemid 20728135)
8. Burrage K,. Parallel and Sequential Methods for Ordinary Differential Equations. Oxford University Press: New York, 1995.
9. Cong NH, Xuan LN,. Twostep-by-twostep PIRK-type PC methods with continuous output formulas. Journal of Computational and Applied Mathematics 2008; 221: 165-173.
10. Nørsett SP, Simonsen HH,. Aspects of parallel Runge-Kutta methods. Numerical Methods for Ordinary Differential Equations (Lecture Notes in Mathematics, vol. 1386), Springer: Berlin, 1989; 103-117.
11. van der Houwen PJ, Sommeijer BP,. Parallel iteration of high-order Runge-Kutta methods with stepsize control. Journal of Computational and Applied Mathematics 1990; 29: 111-127. (Pubitemid 20660003)
12. Schmitt BA, Weiner R, Jebens S,. Parameter optimization for explicit parallel peer two-step methods. Applied Numerical Mathematics 2008; 59: 769-782.
13. Korch M, Rauber T,. Optimizing locality and scalability of embedded Runge-Kutta solvers using block-based pipelining. Journal of Parallel and Distributed Computing 2006; 66 (3): 444-468. (Pubitemid 43227368)
14. Schiesser WE,. The Numerical Method of Lines. Academic Press, Inc.: New York, 1991.
15. Korch M, Rauber T, Scholtes C,. Mixed-parallel implementations of extrapolation methods with reduced synchronization overhead for large shared-memory computers. 2010 IEEE 16th International Conference on Parallel and Distributed Systems (ICPADS), Shanghai, China, 2010; DOI: 10.1109/ICPADS.2010. 12.
16. Korch M, Rauber T, Scholtes C,. Scalability and locality of extrapolation methods for distributed-memory architectures. In Euro-Par 2010-Parallel Processing, D'Ambra P, Guarracino M, Talia D, (eds), (Lecture Notes in Computer Science, vol. 6272). Springer: Berlin/Heidelberg, 2010; 65-76. DOI: 10.1007/978-3-642-15291-7-8.
17. Deuflhard P,. Order and stepsize control in extrapolation methods. Numerical Mathematics 1983; 41: 399-422.
18. Bulirsch R, Stoer J,. Numerical treatment of ordinary differential equations by extrapolation methods. Numerical Mathematics 1966; 8: 1-13.
19. Gragg WB,. On extrapolation algorithms for ordinary initial value problems. SIAM Journal on Numerical Analysis 1965; 2: 384-404.
20. Martí R, Campos V, Piñana E,. A branch and bound algorithm for the matrix bandwidth minimization. European Journal of Operational Research 2008; 186 (2): 513-528. DOI: 10.1016/j.ejor.2007.02.004. (Pubitemid 47615316)
21. Wang Q, Guo YC, Shi XW,. An improved algorithm for matrix bandwidth and profile reduction in finite element analysis. Progress in Electromagnetics Research Letters 2009; 9: 29-38. DOI: 10.2528/PIERL09042305.
22. Korch M, Rauber T,. Parallel low-storage Runge-Kutta solvers for ODE systems with limited access distance. International Journal of High Performance Computing Applications 2010; DOI: 10.1177/1094342010384418.
23. Orozco D, Gao G,. Mapping the FDTD application to many-core chip architectures. The 38th International Conference on Parallel Processing (ICPP-2009). IEEE: New York, 2009.
24. Novell, Optimizing Linux for Dual-Core AMD Opteron Processors, May 2006; Whitepaper.
25. Silicon Graphics, Inc. Linux Application Tuning Guide, January 2009; Document number 007-4639-010.