Abstract
In the context of generating efficient, contention free schedules for inter-node communication through a switch fabric in cluster computing or data center type environments, all-to-all scheduling with equal sized data transfer requests has been studied in the literature [1,3,4]. In this paper, we propose a communication scheduling module (CSM) towards generating contention free communication schedules for many-to-many communication with arbitrary sized data. Towards this end, we propose three approximation algorithms - PST, LDT and SDT. From time to time, the CSM first generates a bipartite graph from the set of received requests, then determines which of these three algorithms gives the best approximation factor on this graph and finally executes that algorithm to generate a contention free schedule. Algorithm PST has a worst case run time of O( max (Δ|E|, |E|log(|E|))) and guarantees an approximation factor of 2H 2Δ− 1, where |E| is the number of edges in the bipartite graph, Δ is the maximum node degree of the bipartite graph and H 2Δ− 1 is the (2Δ− 1)-th harmonic number. LDT runs in O(|E|2) and has an approximation factor of 2(1 + τ), where τ is a constant defined as a guard band or pause time to eliminate the possibility of contention (in an apparently contention free schedule) caused by system jitter and synchronization inaccuracies between the nodes. SDT gives an approximation factor of 4log(w max ) and has a worst case run time of O(Δ|E|log(w max )), where w max represents the longest communication time in a set of received requests.
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
Faraz, A., Patarasuk, P., Yuan, X.: Bandwidth efficient all-to-all broadcast on switched clusters. Intl. Journal of Parallel Programming 36(4), 426–453 (2008)
Fan, X., Jonsson, M., Hoang, H.: Efficient many-to-many real-time communication using an intelligent Ethernet switch. In: Proc. ISPAN, pp. 280–287 (2004)
Tam, A.T.-C., Wang, C.-L.: Contention-aware communication schedule for high-speed communication. Cluster Computing 6(4), 339–353 (2003)
Yang, Y., Wang, J.: Optimal all-to-all personalized exchange in self-routable multistage networks. IEEE Trans. on Parallel and Dist. Systems 11(3), 261–274 (2000)
Aggarwal, G., Motwani, R., Shah, D., Zhu, A.: Switch scheduling via randomized edge coloring. In: Proc. IEEE Symp. on Foundations of Comp. Science (FOCS), p. 502 (2003)
Cole, K.O.R., Schirra, S.: Edge-coloring bipartite multigraphs in O(ElogD) time. Combinatorica 21, 5–12 (2001)
Sanders, P., Steurer, D.: An asymptotic approximation scheme for multigraph edge coloring. ACM Trans. on Algorithms 4(2), 21:1–21:24 (2008)
Firoozshahian, A., Manshadi, V., Goel, A., Prabhakar, B.: Efficient, fully local algorithms for CIOQ switches. In: Proc. INFOCOM, pp. 2491–2495 (2007)
NTP: The Network Time Protocol, http://www.ntp.org/
Operating System Jitter (2010), http://domino.research.ibm.com/comm/research_projects.nsf/pages/osjitter.Identifying.html
Cisco 12000 Series Gigabit Switch Routers (2010), http://www.cisco.com/warp/public/cc/pd/rt/12000/prodlit/gsr_ov.pdf
Touheed, N., et al.: A comparison of dynamic load-balancing algorithms for a parallel adaptive flow solver. Parallel Computing 26(12), 1535–1554 (2000)
Diestel, R.: Graph Theory, 4th edn. Springer, Heidelberg (July 2010)
Sinha, K., Datta Chowdhury, A., Banerjee, S., Ghosh, S.K.: Efficient load balancing on a cluster for large scale online video surveillance. In: Garg, V., Wattenhofer, R., Kothapalli, K. (eds.) ICDCN 2009. LNCS, vol. 5408, pp. 450–455. Springer, Heidelberg (2008)
Gupta, A., Luksch, P., Schmidt, A.C.: MethWerk: scalable mesh based simulation on cluster of SMPs. In: Proc. High Perf. Comp. in Science and Engineering, pp. 141–151 (2005)
Wilcox Jr., T.C.: Dynamic load balancing of virtual machines hosted on Xen, MS Thesis, Dept. of Computer Science, Brigham Young University (April 2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Banerjee, S., Datta Chowdhury, A., Sinha, K., Ghosh, S.K. (2011). Contention-Free Many-to-Many Communication Scheduling for High Performance Clusters. In: Natarajan, R., Ojo, A. (eds) Distributed Computing and Internet Technology. ICDCIT 2011. Lecture Notes in Computer Science, vol 6536. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-19056-8_10
Download citation
DOI: https://doi.org/10.1007/978-3-642-19056-8_10
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-19055-1
Online ISBN: 978-3-642-19056-8
eBook Packages: Computer ScienceComputer Science (R0)