|
Vol: 60(74) No: 2 / June 2015 Self-Organizing Autonomic Control of Distributed, Collaboration Systems through a Leaky Bucket Model Bogdan Solomon NCCT Laboratory, University of Ottawa, School of Electrical Engineering and Computer Science, 161 Louis Pasteur, Room B-306, K1N 6N5, Ottawa, Canada, phone: (613) 562-5800 , e-mail: bsolomon@ncct.uottawa.ca, web: http://www.ncct.uottawa.ca Dan Ionescu NCCT Laboratory, University of Ottawa, School of Electrical Engineering and Computer Science, 161 Louis Pasteur, Room B-306, K1N 6N5, Ottawa, Canada, e-mail: ionescu@ncct.uottawa.ca Cristian Gadea NCCT Laboratory, University of Ottawa, School of Electrical Engineering and Computer Science, 161 Louis Pasteur, Room B-306, K1N 6N5, Ottawa, Canada, e-mail: cris@ncct.uottawa.ca Stejarel Veres NCCT Laboratory, University of Ottawa, School of Electrical Engineering and Computer Science, 161 Louis Pasteur, Room B-306, K1N 6N5, Ottawa, Canada, e-mail: steju@ncct.uottawa.ca Marin Litoiu York University, 4700 Keele Street, M3J 1P3, Toronto, Ontario, Canada, phone: (416) 736-2100 x 209, e-mail: mlitoiu@yorku.ca, web: http://www.yorku.ca/mlitoiu/ Keywords: cloud-computing, performance, SLAs, self-organizing systems Abstract Recently a great deal of research has been undertaken in the area of automating the enterprise IT Infrastructure. For enterprises with a large number of computers the IT Infrastructure represents a considerable amount of the enterprise budget assigned to its operation. Autonomic Computing Systems are systems which were created to correct and optimize the IT infrastructure\'s own self-functioning by executing corrective operations without any need for human intervention. In most cases where autonomic computing systems have been developed this was achieved by the addition of external global controllers. Self-Organizing systems on the other hand are systems which reach a global desired state without the use of any central authority. At the same time the software industry has seen an explosion in the usage of cloud based services. By moving services in the cloud companies can decrease the IT budget such that they only use the resources which are needed, when they are needed. In this paper, an autonomic system based on a self-organizing architecture and model is introduced. The system is applied to the self-optimization of a web-based real-time collaboration application running on a geographically distributed cloud. The model is based on the leaky bucket model commonly used in network control, while the control system is a self-organizing system. A testbed for the architecture is also presented and used in order to gather performance data from the collaboration application. References [1] B. Solomon, D. Ionescu, C. Gadea, S. Veres, M. Litoiu, and J. Ng, “Distributed clouds for collaborative applications,” in Collaboration Technologies and Systems (CTS), 2012 International Conference on, May 2012, pp. 218 – 225. [2] B. Solomon, D. Ionescu, M. Litoiu, and G. Iszlai, “Self-organizing autonomic computing systems,” in Logistics and Industrial Informatics (LINDI), 2011 3rd IEEE International Symposium on, aug. 2011, pp. 99 – 104. [3] V. Guffens and G. Bastin, “Optimal adaptive feedback control of a network buffer,” in American Control Conference, 2005. Proceedings of the 2005, vol. 3, June 2005, pp. 1835 – 1840. [4] V. Guffens, G. Bastin, and H. Mounier, “Using token leaky buckets for congestion feedback control in packet switched networks with guaranteed boundedness of buffer queues,” in Proceedings of European Control Conference (ECC), 2003. [5] OpenStack, “OpenStack Cloud Software,” OpenStack, [Accessed: January 2013]. [Online]. Available: http://www.openstack.org/. [6] Red5, “Red5 Media Server,” Red5, [Accessed: January 2013]. [Online]. Available: http://www.red5.org/. [7] L. Deri. ntop. ntop. [Accessed: January 2013]. [Online]. Available: http://www.ntop.org/ [8] B. Solomon, D. Ionescu, C. Gadea, S. Veres and M. Litoiu. “Leaky Bucket Model for Autonomic Control of Distributed, Collaborative Systems”, in SACI 2013. [9] R.-E. Precup and S. Preitl, “Popov-type stability analysis method for fuzzy control systems,” in Proc. Fifth European Congress on Intelligent Technologies and Soft Computing (EUFIT’97), Aachen, Germany, 1997, vol. 2, pp. 1306–1310. |