Abstract
The latency of network switching is of particular importance for fine-grained communication processes such as memory access. We propose a specialized network switch that reduces communication latency because many scientific computing applications have specific access patterns. In this paper, we describe the basic concept and design of our reconfigurable shared memory network. Our evaluation results show that there is no significant difference between the port-to-port latency of our network switch and that of an 8-port InfiniBand network switch. Furthermore, the hardware size of the routing stack in the network switch does not increase with the number of nodes.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Chen, W.Y., Iancu, C., Yelick, K.: Communication optimizations for fine-grained upc applications. In: Proceedings of the 14th International Conference on Parallel Architectures and Compilation Techniques, PACT 2005, pp. 267–278. IEEE Computer Society (2005)
Dandapanthula, N., Subramoni, H., Vienne, J., Kandalla, K., Sur, S., Panda, D.K., Brightwell, R.: INAM - A Scalable InfiniBand Network Analysis and Monitoring Tool. In: Alexander, M., D’Ambra, P., Belloum, A., Bosilca, G., Cannataro, M., Danelutto, M., Di Martino, B., Gerndt, M., Jeannot, E., Namyst, R., Roman, J., Scott, S.L., Traff, J.L., Vallée, G., Weidendorfer, J. (eds.) Euro-Par 2011, Part II. LNCS, vol. 7156, pp. 166–177. Springer, Heidelberg (2012)
Fillo, M., Gillett, R.B.: Architecture and implementation of memory channel 2. Digital Tech. J. 9, 27–41 (1997)
Hoefler, T., Schneider, T., Lumsdaine, A.: Multistage switches are not crossbars: Effects of static routing in high-performance networks. In: 2008 IEEE International Conference on Cluster Computing, October 1-29, pp. 116–125 (2008)
The Xcalable MP: Directive-based language extension for scalable and performance-aware parallel programming, http://www.xcalablemp.org/
The Berkeley UPC Compiler (2002), http://upc.lbl.gov/
Matrix Market (2007), http://math.nist.gov/MatrixMarket/
Kamil, S., Pinar, A., Gunter, D., Lijewski, M., Oliker, L., Shalf, J.: Reconfigurable hybrid interconnection for static and dynamic scientific applications. In: Proceedings of the 4th International Conference on Computing Frontiers, CF 2007, pp. 183–194. ACM (2007)
Mellanox InfiniBand switches (2012), http://www.mellanox.com/
Ohnishi, Y., Yoshida, T.: Design and evaluation of a distributed shared memory network for application-specific pc cluster systems. In: Proceedings of the 2011 IEEE Workshops of International Conference on Advanced Information Networking and Applications, WAINA 2011, pp. 63–70. IEEE Computer Society (2011)
Tanabe, N., Yamamoto, J., Nishi, H., Kudoh, T., Hamada, Y., Nakajo, H., Amano, H.: Low latency high bandwidth message transfer mechanisms for a network interface plugged into a memory slot. Cluster Computing 5(1), 7–17 (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ohnishi, Y., Yoshida, T. (2012). Design of an Application-Dependent Static-Based Shared Memory Network. In: Xiang, Y., Stojmenovic, I., Apduhan, B.O., Wang, G., Nakano, K., Zomaya, A. (eds) Algorithms and Architectures for Parallel Processing. ICA3PP 2012. Lecture Notes in Computer Science, vol 7440. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-33065-0_7
Download citation
DOI: https://doi.org/10.1007/978-3-642-33065-0_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-33064-3
Online ISBN: 978-3-642-33065-0
eBook Packages: Computer ScienceComputer Science (R0)