Porting scientific libraries to PGAS in XSEDE resources: practice and experience

作者: Antonio Gómez-Iglesias , Dmitry Pekurovsky , Khaled Hamidouche , Jie Zhang , Jérôme Vienne

DOI: 10.1145/2792745.2792785

关键词:

摘要: The next generation of supercomputers presents new and complex challenges that might require a change in the current paradigm how parallel applications are developed. Hybrid programming is usually described as best approach for exascale computers. PGAS models considered an interesting alternative to work together with MPI this hybrid model achieve good performance those machines. This very promising especially one-sided irregular communication patterns. However, still emerging technology there not much previous experience on port existing model. Due relevance devices, it relevant have early porting well knowledge issues be faced paradigm. In paper we present two different scientific currently implemented candidates We describe these been ported, some solutions found. also show can great when compared MPI.

参考文章(19)
Torsten Hoefler, Jeffrey M. Squyres, Wolfgang Rehm, Andrew Lumsdaine, A Case for Non-blocking Collective Operations Frontiers of High Performance Computing and Networking – ISPA 2006 Workshops. pp. 155- 164 ,(2006) , 10.1007/11942634_17
Ayon Basumallik, Rudolf Eigenmann, Optimizing irregular shared-memory applications for distributed-memory systems acm sigplan symposium on principles and practice of parallel programming. pp. 119- 128 ,(2006) , 10.1145/1122971.1122990
Jerome Vienne, Jitong Chen, Md. Wasi-Ur-Rahman, Nusrat S. Islam, Hari Subramoni, Dhabaleswar K. Panda, Performance Analysis and Evaluation of InfiniBand FDR and 40GigE RoCE on HPC and Cloud Computing Systems 2012 IEEE 20th Annual Symposium on High-Performance Interconnects. pp. 48- 55 ,(2012) , 10.1109/HOTI.2012.19
Dmitry Pekurovsky, P3DFFT: A Framework for Parallel Computations of Fourier Transforms in Three Dimensions SIAM Journal on Scientific Computing. ,vol. 34, ,(2012) , 10.1137/11082748X
Jiuxing Liu, Jiesheng Wu, Sushmitha P. Kini, Pete Wyckoff, Dhabaleswar K. Panda, High performance RDMA-based MPI implementation over InfiniBand Proceedings of the 17th annual international conference on Supercomputing - ICS '03. ,vol. 32, pp. 295- 304 ,(2003) , 10.1145/782814.782855
H Homann, O Kamps, R Friedrich, R Grauer, Bridging from Eulerian to Lagrangian statistics in 3D hydro- and magnetohydrodynamic turbulent flows New Journal of Physics. ,vol. 11, pp. 073020- ,(2009) , 10.1088/1367-2630/11/7/073020
Jörg Schumacher, Lagrangian studies in convective turbulence Physical Review E. ,vol. 79, pp. 056301- ,(2009) , 10.1103/PHYSREVE.79.056301
Robert Preissl, Nathan Wichmann, Bill Long, John Shalf, Stephane Ethier, Alice Koniges, Multithreaded Global Address Space Communication Techniques for Gyrokinetic Fusion Applications on Ultra-Scale Platforms Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis on - SC '11. pp. 78- ,(2011) , 10.1145/2063384.2071033
Christopher S. Simmons, Karl W. Schulz, A distributed memory out-of-core method on HPC clusters and its application to quantum chemistry applications Proceedings of the 1st Conference of the Extreme Science and Engineering Discovery Environment on Bridging from the eXtreme to the campus and beyond - XSEDE '12. pp. 1- ,(2012) , 10.1145/2335755.2335785
Mingzhe Li, Jian Lin, Xiaoyi Lu, Khaled Hamidouche, Karen Tomko, Dhabaleswar K Panda, None, Scalable MiniMD Design with Hybrid MPI and OpenSHMEM Proceedings of the 8th International Conference on Partitioned Global Address Space Programming Models. pp. 24- ,(2014) , 10.1145/2676870.2676893