Abstract
This paper presents an approach for service discovery and service sharing in an auto-load balanced cluster environment. We propose a B+ tree based architectural framework for dynamic configuration of cluster to achieve load balancing. Our prime focus is on the discovery and sharing of services in the cluster in auto-configurable load balanced cluster. The load balancing is required because there are instances when a node is burdened with too many connections accessing a particular service. This demands replication of service to an under loaded node so as to optimize the cluster performance. The implementation of proposed approach is carried out by using MOSIX as a middleware. Experimental results and simulations shows that the configuration cost of a cluster size of 50-60 nodes have lowest execution time of 2200 microseconds, which is significantly better than other approaches. Our mechanism is able to show that replication of service on the idle nodes of local or same cluster keeps the rate of traffic significantly low, which in turn increases the overall performance of the cluster.
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
Preview
Unable to display preview. Download preview PDF.
References
Liu, P.-W., Chan, L.-S., Ling, F.-Y., Hang, S.-S.: Distributed Computing: New Power for Scientific Visualization 16(3) (May 1996)
Buyya, R.: High Performance Cluster Computing. Pearson Education 1, 519–532 (2008)
Kozat, U., Tassiula, L.: Service discovery in mobile adhoc Networks: an Overall Perspective on architectural choices and network Layer support issues. Adhoc Networks 2(1), 23–44 (2004)
Safa, H., Artail, H., Hamze, H., Mershad, K.: A collaborative service discovery and service sharing framework for mobile ad hoc networks. In: Li, K., Jesshope, C., Jin, H., Gaudiot, J.-L. (eds.) NPC 2007. LNCS, vol. 4672, pp. 151–160. Springer, Heidelberg (2007)
Milojicic, D., HP Labs Douglis, F.: Process Migration Review. AT&T Labs-Research,Yves Paindaveine TOG research Institute, Richard Wheeler EMC and Songnain Zhou University of Toronto and Platform Computing published in ACM Computing Surveys 32(3), 241–299 (2000)
Barak, A., Shiloh, A.: The MOSIX Management System for Linux Clusters, Multi-Clusters and Clouds. White Paper
Caron, B., Garonne, V., Tsaregorodtsev, A.: Definition, modeling and simulation of a cluster computing scheduling system for high throughput computing. Future Generation Computer Systems 23, 968–976 (2007)
Service Oriented Architecture (SOA) and Specialized Messaging Patterns Technical white paper by Adobe
Yefim, V., Gartner, N.: Service-Oriented Architecture Scenario. publication ID Number: AV-19-6751 (2003)
Li, Y., Lan, Z.: A Survey of Load Balancing in Cluster Computing. Department of Computer Science, Illinois Institute of Technology, Chicago, IL 60616
Mishra, S., Kushwaha, D.S., Misra, A.K.: Hybrid Load Balancing in Auto-Configurable Load Balanced Cluster. Journal of Computer Science & Engineering 2(1), 16–25 (2010)
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
Mishra, S., Mathur, A., Agarwal, H., Vashishtha, R., Kushwaha, D.S., Misra, A.K. (2011). Discovery and High Availability of Services in Auto-load Balanced Clusters. In: Meghanathan, N., Kaushik, B.K., Nagamalai, D. (eds) Advances in Computer Science and Information Technology. CCSIT 2011. Communications in Computer and Information Science, vol 131. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-17857-3_48
Download citation
DOI: https://doi.org/10.1007/978-3-642-17857-3_48
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-17856-6
Online ISBN: 978-3-642-17857-3
eBook Packages: Computer ScienceComputer Science (R0)