Abstract
Fat-trees (FTs) are widely known topologies that, among other advantages, provide full bisection bandwidth. However, many implementations of FTs are made slimmed to cheapen the infrastructure, since most applications do not make use of this full bisection bandwidth. In this paper Extended Generalized Random Folded Clos (XGRFC) interconnection networks are introduced as cost-efficient alternatives to Extended Generalized Fat Trees (XGFT), which is a widely used topological description for slimmed FTs. This is proved both by obtaining a theoretical model of the performance and evaluating it using simulation. Among the results, it is shown that a XGRFC is able to connect 20k servers with 27% less routers than the corresponding XGFT and still providing the same performance under uniform traffic.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Al-Fares, M., Loukissas, A., Vahdat, A.: A scalable, commodity data center network architecture. In: Proceedings of the ACM SIGCOMM 2008 Conference on Data Communication, SIGCOMM 2008, pp. 63–74. ACM, New York (2008)
Camarero, C., Martínez, C., Beivide, R.: On random wiring in practicable folded clos networks for modern datacenters. IEEE Trans. Parallel Distrib. Syst. 29(8), 1780–1793 (2018). https://doi.org/10.1109/TPDS.2018.2805344
Camarero, C., Martínez, C., Beivide, R.: Random folded Clos topologies for datacenter networks. In: Proceedings of the 23rd IEEE Symposium on High Performance Computer Architecture, HPCA 2017, pp. 193–204 (2017)
Chen, D., Eisley, N., Heidelberger, P., et al.: Looking under the hood of the IBM Blue Gene/Q network. In: Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, SC 2012, pp. 69:1–69:12. IEEE Computer Society Press, Los Alamitos (2012)
Clos, C.: A study of non-blocking switching networks. Bell Syst. Tech. J. 32(2), 406–424 (1953)
Jain, R., Chiu, D., Hawe, W.: A quantitative measure of fairness and discrimination for resource allocation in shared computer systems. In: DEC Research Report TR-301, pp. 598–607, June 1984
Jokanovic, A., Sancho, J.C., Labarta, J., Rodriguez, G., Minkenberg, C.: Effective quality-of-service policy for capacity high-performance computing systems. In: 2012 IEEE 14th International Conference on High Performance Computing and Communication, 2012 IEEE 9th International Conference on Embedded Software and Systems, pp. 598–607, June 2012
Kamil, S., Oliker, L., Pinar, A., Shalf, J.: Communication requirements and interconnect optimization for high-end scientific applications. IEEE Trans. Parallel Distrib. Syst. 21(2), 188–202 (2010)
León, E.A., Karlin, I., Bhatele, A., et al.: Characterizing parallel scientific applications on commodity clusters: an empirical study of a tapered fat-tree. In: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, pp. 909–920, November 2016
Navaridas, J., Miguel-Alonso, J., Pascual, J.A., Ridruejo, F.J.: Simulating and evaluating interconnection networks with INSEE. Simul. Model. Pract. Theor. 19(1), 494–515 (2011)
Ohring, S.R., Ibel, M., Das, S.K., Kumar, M.J.: On generalized fat trees. In: Proceedings of 9th International Parallel Processing Symposium, pp. 37–44, April 1995
Rodríguez, G.: Understanding and reducing contention in generalized fat tree networks for high performance computing. Ph.D. thesis, Facultad de Informática, U. Politécnica de Cataluña (2011)
Shpiner, A., Haramaty, Z., Eliad, S., Zdornov, V., Gafni, B., Zahavi, E.: Dragonfly+: low cost topology for scaling datacenters. In: 2017 IEEE 3rd International Workshop on High-Performance Interconnection Networks in the Exascale and Big-Data Era (HiPINEB), pp. 1–8 (2017)
Acknowledgements
This work has been supported by the Spanish Ministry of Science, Innovation and Universities under contract TIN2016-76635-C2-2-R (AEI/FEDER, UE). The first author is supported by the Spanish Minister of Science and Innovation programme Juan del Cierva Formación reference FJCI-2017-31643.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Camarero, C., Corral, J., Martínez, C., Beivide, R. (2020). Modelling Standard and Randomized Slimmed Folded Clos Networks. In: Malawski, M., Rzadca, K. (eds) Euro-Par 2020: Parallel Processing. Euro-Par 2020. Lecture Notes in Computer Science(), vol 12247. Springer, Cham. https://doi.org/10.1007/978-3-030-57675-2_12
Download citation
DOI: https://doi.org/10.1007/978-3-030-57675-2_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-57674-5
Online ISBN: 978-3-030-57675-2
eBook Packages: Computer ScienceComputer Science (R0)