Today Publish-subscribe model is used as communication backbone for various application domains such as IoT, Social networking, Intrusion detection system and Financial trading. Content-based flavor of Pub-Sub system enables routing of information from producers to consumers based on contents of the query or depends on subscriptions entered by the user. In this model, information is disseminated from producers to consumers through a network of brokers. The significant challenge in content-based Pub-Sub system lies in an efficient matching of an event against a large number of subscribers on a single message broker. To provide high throughput service guarantee to the subscriber of Pub-Sub system we propose a novel hybrid model for parallel event processing using MPI-CUDA approach. This approach combines message passing interface (MPI) and CUDA, a parallel computing platform and programming model, invented by NVIDIA. Results are compared with CCM (Cuda Content Matching Algorithm), a high performance, and parallel content matching algorithm. Approximately 1.77X speedup is observed in matching latency. This approach is suitable for use in event processing of large data intensive applications where the rate of arrival of the event is high.

Matching Latency, MPI-CUDA, High performance, parallel event processing.

[1] Alessandro Margara, Gianpaolo Cugola “High-Performance Publish-Subscribe Matching Using Parallel Hardware”, IEEE Transactions on Parallel and Distributed Systems Volume.25, Issue.1,January 2014
[2] A. Carzaniga, D. S. Rosenblum, and A. L. Wolf, “Design and evaluation of a wide-area event notification service”, ACM TCS, 2001
[3] M. K. Aguilera, R. E. Strom, D. C. Sturman, M. Astley, and T. D. Chandra, “Matching events in a content-based subscription system”, In PODC, 1999 and evaluation of a wide-area event notification service. ACM TCS, 2001.
[4] H.-A. Jacobsen, A. Cheung, G. Lia, B. Maniymaran, V. Muthusamy, and R. S. Kazemzadeh, “The PADRES publish/subscribe system”, Handbook of Research on Adv. Dist. Event-Based Sys. Pub. /Sub. and Message Filtering Tech., 2009.
[5] Peter R. Pietzuch, “Hermes: A scalable event-based middleware”, Technical Report University of GANBRIDGG Computer Laboratory.
[6] T. Yan and H. Garcia-Molina, “Index structures for selective dissemination of information under the Boolean model”, ACM TODS,1994.
[7] F. Fabret, H.-A. Jacobsen, F. Llirbat, J. Pereira, K. A. Ross, and D. Shasha, “Filtering algorithms and implementation for fast pub/sub systems”, SIGMOD, 2001.
[8] S. Whang, C. Brower, J. Shanmugasundaram, S. Vassilvitskii, E. Veer, R. Yerneni, and H. Garcia-Molina, “Indexing Boolean expressions”, In VLDB’09.
[9] M. K. Aguilera, R. E. Strom, D. C. Sturman, M. Astley, and T. D. Chandra, “Matching events in a content-based subscription system”, In PODC, 1999.
[10] A. Campailla, S. Chaki, E. Clarke, S. Jha, and H. Veith, “Efficient filtering in publish-subscribe systems using binary decision diagrams”, In ICSE, 2001.
[11] G. Li, S. Hou, and H.-A. Jacobsen, “A unified approach to routing, covering and merging in publish/subscribe systems based on modified binary decision diagrams”, In ICDCS,2003.
[12] G. Cugola and G. Picco, “REDS: A Reconfigurable Dispatching System”, In SEM, pages 9-16, Portland, 2006. ACM Press.

[13] Tania Banerjee Mishra, Sartaj Sahni, “PUBSUB: An Efficient Publish/Subscribe System”, IEEE Transactions on Computers Volume. 64, Issue.4, 2014.
[14] Mohammad Sadoghi, Hans-Arno Jacobsen, “Analysis and Optimization for Boolean Expression Indexing ACM Transactions on Database Systems”, Vol. 38, No. 2, Article 8, 2013.
[15] M. Sadoghi and H.-A. Jacobsen, “BE-Tree An Index Structure to Efficiently Match Boolean Expressions over High-dimensional Discrete Space”, SIGMOD, 2011.
[16] A. Farroukh, E. Ferzli, N. Tajuddin, and H.-A. Jacobsen, “Parallel Event Processing for Content-Based Publish/Subscribe Systems”, (DEBS ’09), pp.8:1-8:4, 2009.
[17] K.H. Tsoi, I. Papagiannis, M. Migliavacca, W. Luk, and P Pietzuch, “Accelerating Publish/Subscribe Matching on Reconfigurable Supercomputing Platforms”, Proc. Many-Core and Reconfigurable Supercomputing Conf., 2010.
[18] Medha A. Shah, D.B.Kulkarni, “Storm Pub-Sub: High Performance, Scalable Content Based Event Matching System Using Storm”, In IPDPS,W’ 2015.
[19] Raphael Barazzutti, Pascal Felber, “Streamhub: A Massively Parallel Architecture for High-Performance Content-Based Publish/Subscribe”, DEBS, 2013, Arlington, Texas, USA.
[20] Medha Shah, D.B.Kulkarni, “Enabling Qos Support for Multi-Core Message Broker in Publish/Subscribe System”, Advance Computing Conference (IACC), 2014 IEEE International conference.
[21] Zhaoran Wang, Xiaotao Chang, “Pub/Sub on Stream: A Multi-Core Based Message Broker with Qos Support”, DEBS, 2012, Berlin, Germany July 2012.Forman, G. 2003. An extensive empirical study of feature selection metrics for text classification. J. Mach. Learn. Res. 3 (Mar. 2003), 1289-1305.
[22] A. Carzaniga and A.L. Wolf, “Forwarding in a Content-Based Network,” Proc. SIGCOMM, pp. 163-174, 2003.
[23] https://www.linux.com/blog/building-beowulf-cluster-just-13-steps