Abstract
MPI has been used as a parallel programming model for supercomputers and clusters and recently in MultiProcessor Systems-on-Chip (MPSoC). One component of MPI is collective communication and its performance is key for certain parallel applications to achieve good speedups. Previous work showed that, with synthetic communication-only benchmarks, communication improvements of up to 11.4-fold and 22-fold for broadcast and reduce operations, respectively, can be achieved by providing hardware support at the network level in a Network-on-Chip (NoC). However, these numbers do not provide a good estimation of the advantage for actual applications, as there are other factors that affect performance besides communications, such as computation. To this end, we extend our previous work by evaluating the impact of hardware support over a set of five parallel application kernels of varying computation-to-communication ratios. By introducing some useful computation to the performance evaluation, we obtain more representative results of the benefits of adding hardware support for broadcast and reduce operations. The experiments show that applications with lower computation-to-communication ratios benefit the most from hardware support as they highly depend on efficient collective communications to achieve better scalability. We also extend our work by doing more analysis on clock frequency, resource usage, power, and energy. The results show reasonable scalability for resource utilization and power in the network interfaces as the number of channels increases and that, even though more power is dissipated in the network interfaces due to the added hardware, the total energy used can still be less if the actual speedup is sufficient. The application kernels are executed in a 24-embedded-processor system distributed across four FPGAs.
- L. A. Aguilar, D. A. Steinman, and R. S. C. Cobbold. 2010. On the synthesis of sample volumes for real-time spectral doppler ultrasound simulation. Ultrasound Med. Biol. 36, 12, 2107--2116.Google Scholar
Cross Ref
- Q. Ali, S. P. Midkiff, and V. S. Pai. 2009. Efficient high performance collective communication for the cell blade. In Proceedings of the 23rd International Conference on Supercomputing (ICS'09). ACM Press, New York, 193--203. Google Scholar
Digital Library
- M. P. Allen and D. J. Tildesley. 1987. Computer Simulation of Liquids. Clarendon Press, New York. Google Scholar
Digital Library
- G. Almasi, P. Heidelberger, C. J. Archer, X. Martorell, C. C. Erway, J. E. Moreira, B. Steinmacher-Burow, and Y. Zheng. 2005. Optimization of mpi collective communication on bluegene/l systems. In Proceedings of the 19th Annual International Conference on Supercomputing (ICS'05). ACM Press, New York, 253--262. Google Scholar
Digital Library
- M. Barnett, R. Littlefield, D. Payne, and R. Van De Geijn. 1993. Global combine on mesh architectures with wormhole routing. In Proceedings of the 7th International Parallel Processing Symposium. 156--162. Google Scholar
Digital Library
- Beecube 2011. Beecube. http://beecube.com/.Google Scholar
- I. S. Dhillon and D. S. Modha. 2000. A data-clustering algorithm on distributed memory multiprocessors. In Proceedings of the Revised Papers from Large-Scale Parallel Data Mining, Workshop on Large-Scale Parallel KDD Systems (SIGKDD'00). Springer, 245--260. Google Scholar
Digital Library
- S. Gao, A. Schmidt, and R. Sass. 2010. Impact of reconfigurable hardware on accelerating MPI reduce. In Proceedings of the International Conference on Field-Programmable Technology (FPT'10). 29--36.Google Scholar
- T. Hoefler, C. Siebert, and W. Rehm. 2007. A practically constant-time MPI broadcast algorithm for large-scale infiniband clusters with multicast. In Proceedings of the IEEE International Parallel and Distributed Processing Symposium (IPDPS'07). 1--8.Google Scholar
- J. Liu, A. R. Mamidala, and D. K. Panda. 2003. Fast and scalable MPI-level broadcast using infiniband's hardware multicast support. In Proceedings of the IEEE International Parallel and Distributed Processing Symposium (IPDPS'07).Google Scholar
- P. Mahr, C. Lorchner, H. Ishebabi, and C. Bobda. 2008. SoC-MPI: A flexible message passing library for multiprocessor systems-on-chips. In Proceedings of the International Conference on Reconfigurable Computing and FPGAs. IEEE Computer Society, 187--192. Google Scholar
Digital Library
- MPI Forum. 1993. MPI: A message passing interface. In Proceedings of the ACM/IEEE Conference on Supercomputing (Supercomputing'93). ACM Press, New York, 878--883. Google Scholar
Digital Library
- P. S. Pacheco. 1997. An application: Numerical integration. In Parallel Programming with MPI, Morgan Kaufmann Publishers, San Francisco, 53--60.Google Scholar
- Y. Peng, M. Saldana, and P. Chow. 2011. Hardware support for broadcast and reduce in MPSOC. In Proceedings of the 21st International Conference on Field-Programmable Logic and Applications. 144--150. Google Scholar
Digital Library
- M. Saldana, A. Patel, C. Madill, D. Nunes, D. Wang, P. Chow, R. Wittig, H. Styles, and A. Putnam. 2010. MPI as a programming model for high-performance reconfigurable computers. ACM Trans. Reconfig. Technol. Syst. 3, 22:1--22:29. Google Scholar
Digital Library
- K. D. Underwood, W. B. Ligon III, and R. R. Sass. 2003. Analysis of a prototype intelligent network interface. Concurr. Comput. Pract. Exper. 15, 7--8, 751--777.Google Scholar
Cross Ref
- M. K. Velamati, A. Kumar, N. Jayam, G. Senthilkumar, P. K. Baruah, R. Sharma, S. Kapoor, and A. Srinivasan. 2007. Optimization of collective communication in intra-cell MPI. In Proceedings of the 14th International Conference on High-Performance Computing (HiPC'07). Springer, 488--499. Google Scholar
Digital Library
- Voltaire. 2011. Voltaire. http://www.voltaire.com/.Google Scholar
- Xpower. 2011. Xilinx. http://www.xilinx.com/.Google Scholar
- J. Zhu. 1994. Solving Partial Differential Equations on Parallel Computers. World Scientific. Google Scholar
Digital Library
Index Terms
Benefits of Adding Hardware Support for Broadcast and Reduce Operations in MPSoC Applications
Recommendations
Hardware Support for Broadcast and Reduce in MPSoC
FPL '11: Proceedings of the 2011 21st International Conference on Field Programmable Logic and ApplicationsMPI has been used as a parallel programming model for supercomputers and clusters but also in Multiprocessor System-on-Chip. One component of MPI is collective communication and its performance is key for parallel applications to achieve good speedups. ...
Coprocessor design to support MPI primitives in configurable multiprocessors
The Message Passing Interface (MPI) is a widely used standard for interprocessor communications in parallel computers and PC clusters. Its functions are normally implemented in software due to their enormity and complexity, thus resulting in large ...
A Detailed Performance Analysis of the Interpolation Supplemented Lattice Boltzmann Method on the Cray T3E and Cray X1A Detailed Performance Analysis of the Interpolation Supplemented Lattice Boltzmann Method on the Cray T3E and Cray X1
A detailed study of the parallel performance of the interpolation supplemented lattice Boltzmann (ISLB) method using SHMEM and MPI on the Cray T3E-900 and Cray X1 architectures is presented. The noteworthy feature of the ...






Comments