Abstract
New advances in reconfigurable optical interconnect technologies will allow the fabrication of low-cost, fast and run-time adaptable networks for connecting processors and memory modules in large distributed shared-memory multiprocessor machines. Since the switching times of these components are typically high compared to the memory access time, reconfiguration exploits low frequency dynamics in the network traffic patterns. These are however not easily reproduced using statistical traffic generation, a tool commonly used when doing a fast design space exploration. In this paper, we present a technique that can predict network performance, based on the traffic patterns obtained from simulating the execution of real benchmark applications, but without the need to perform these slow full-system simulations for every parameter set of interest. This again allows for a quick comparison of different network implementations with good relative accuracy, narrowing down the design space for more detailed examination.
Similar content being viewed by others
References
Miller D.A.B. and Ozaktas H.M. (1997). Limit to the bit-rate capacity of electrical interconnects from the aspect ratio of the system architecture. J. Parallel Distributed Comput. 41(1): 42–52
Lenoski D., Laudon J., Gharachorloo K., Weber W.-D., Gupta A., Hennessy J.L., Horowitz M. and Lam M.S. (1992). The Stanford DASH multiprocessor. IEEE Comput. 25(3): 63–79
Collet J., Litaize D., Campenhout J.V., Desmulliez M., Jesshope C., Thienpont H., Goodman J. and Louri A. (2000). Architectural approach to the role of optics in monoprocessor and multiprocessor machines. Appl. Opt. 39(5): 671–682
Benner A.F., Ignatowski M., Kash J.A., Kuchta D.M. and Ritter M.B. (2005). Exploitation of optical interconnects in future server architectures. IBM J. Res. Dev. 49(4/5): 755–776
Mohammed E. (2004). Optical interconnect system integration for ultra-short-reach applications. Intel Technol. J. 8(2): 115–127
Schares, L., et al.: Terabus—a waveguide-based parallel optica interconnect for Tb/s-class on-board data transfers in computer systems. In: Proceedings of the 31st European Conference on Optical Communication (ECOC 2005), vol. 3, pp. 369–372. The Institution of Electrical Engineers, Glasgow, Scotland (2005)
Heirman, W., Artundo, I., Desmet, L., Dambre, J., Debaes, C., Thienpont, H., Van Campenhout, J.: Speeding up multiprocessor machines with reconfigurable optical interconnects. In: Eldada, L., Lee, E.-H. (eds.) Proceedings of SPIE, Optoelectronic Integrated Circuits VIII, Photonics West, vol. 6124, p. 61240K. SPIE, San Jose, California, USA (2006)
Katsinis C. (2001). Performance analysis of the simultaneous optical multi-processor exchange bus. Parallel Comput. 27(8): 1079–1115
Heirman, W., Dambre, J., Van Campenhout, J.,Debaes, C., Thienpont, H.: Traffic temporal analysis for reconfigurable interconnects in shared-memory systems. In: Proceedings of the 19th IEEE International Parallel & Distributed Processing Symposium, IEEE Computer Society, p. 150. Denver, Colorado (2005)
Magnusson P.S., Christensson M., Eskilson J., Forsgren D., Hallberg G., Hogberg J., Larsson F., Moestedt A. and Werner B. (2002). Simics: A full system simulation platform. IEEE Comput. 35(2): 50–58
Ridruejo, F., Gonzalez, A., Miguel-Alonso, J.: TrGen: a traffic generation system for interconnection network simulators. In: 1st. Int. Workshop on Performance Evaluation of Networks for Parallel, Cluster and Grid Computing Systems (PEN-PCGCS’05), pp. 547–553. Olso, Norway (2005)
Brunfaut, M., Meeus, W., Van Campenhout, J., Annen, R., Zenklusen, P., Melchior, H., Bockstaele, R., Vanwassenhove, L., Hall, J., Wittman, B., Nayer, A., Heremans, P., Van Koetsem, J., King, R., Thienpont, H., Baets, R.: Demonstrating optoelectronic interconnect in a FPGA based prototype system using flip chip mounted 2D arrays of optical components and 2D POF-ribbon arrays as optical pathways. In: Proceedings of SPIE, vol. 4455, pp. 160–171. Bellingham (2001).
Huang D., Sze T., Landin A., Lytel R. and Davidson H. (2003). Optical interconnects: out of the box forever?. IEEE J. Select. Top. Quant. Electron. 9(2): 614–623
Snyder L. (1982). Introduction to the configurable, highly parallel computer. Computer 15(1): 47–56
Pinkston T.M. and Goodman J.W. (1994). Design of an optical reconfigurable shared-bus-hypercube interconnect. Appl. Opt. 33(8): 1434–1443
Han X., Chen R.T. (2004) Improvement of multiprocessing performance by using optical centralized shared bus. In: Proceedings of the SPIE, vol. 5358, pp. 80–89 (2004)
Shacham A., Small B.A., Liboiron-Ladouceur O. and Bergman K. (2005). A fully implemented 12 × 12 data vortex optical packet switching interconnection network. J. Lightw. Technol. 23(10): 3066–3075
Artundo I., Desmet L., Heirman W., Debaes C., Dambre J., Van Campenhout J. and Thienpont H. (2006). Selective optical broadcast component for reconfigurable multiprocessor interconnects. IEEE J. Select. Topic. Quant. Electron.: Spec. Issue Opt. Commun. 12(4): 828–837
Heirman W., Dambre J., Artundo I., Debaes C., Thienpont H., Stroobandt D. and Van Campenhout J. (2007). Predicting reconfigurable interconnect performance in distributed shared-memory systems. Integ. the VLSI J. 40(4): 382–393
Sterling, T., Savarese, D., Becker, D.J., Dorband, J.E., Ranawake, U.A., Packer, C.V.: Beowulf: A parallel workstation for scientic computation, In: Proceedings of the International Conference on Parallel Processing, pp. 11–14. CRC Press, Boca Raton, USA (1995)
Sánchez, J.L., Duato, J., García, J. M.: Using channel pipelining in reconfigurable interconnection networks. In: 6th Euromicro Workshop on Parallel and Distributed Processing, 1998.
Leiserson C.E., Abuhamdeh Z.S., Douglas D.C., Feynman C.R., Ganmukhi M.N., Hill J.V., Hillis W.D., Kuszmaul B.C., Pierre M.A.S., Wells D.S., Wong-Chan M.C., Yang S.-W. and Zak R. (1996). The network architecture of the Connection Machine CM-5. J. Parallel Distributed Comput. 33(2): 145–158
Woo, S.C., Ohara, M., Torrie, E., Singh, J.P., Gupta, A.: The SPLASH-2 programs: Characterization and methodological considerations. In: Proceedings of the 22th International Symposium on Computer Architecture, pp. 24–36. Santa Margherita Ligure, Italy (1995).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Heirman, W., Dambre, J., Artundo, I. et al. Predicting the performance of reconfigurable optical interconnects in distributed shared-memory systems. Photon Netw Commun 15, 25–40 (2008). https://doi.org/10.1007/s11107-007-0084-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11107-007-0084-z