Loading

Reliability Enhancement from http Log Files In Composite Web Services
P. Maruthurkarasi (alias) Rohini1, C. Jayaprakash2, R. Balaji Ganesh3
1P.Maruthurkarasi (alias) Rohini, Department of Information Technology, K.L.N. College of Information Technology, Pottapalayam, Sivagangai.
2C.Jayaprakash, Associate Professor, Department of Information Technology, K.L.N. College of Information Technology, Pottapalayam, Sivagangai.
3R. Balaji Ganesh, Assistant Professor, Department of Information Technology, K.L.N. College of Information Technology, Pottapalayam, Sivagangai.
Manuscript received on March 20, 2013. | Revised Manuscript received on April 13, 2013. | Manuscript published on April 30, 2013. | PP: 464-468 | Volume-2, Issue-4, April 2013. | Retrieval Number: D1473042413/2013©BEIESP

Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: A Hybrid Reliability Model based on log analyzer is designed to evaluate the Reliability of Composite Web Services. Based on Dependability, Atomic Web Services are composed with a Central Co-ordination Function (Broker). Real Time Server Log Files are fed as input to the system. Log Analyzer reads the log entries and separates the individual response of the server along with the time stamps. Base on the Frequency of the service response are classified and the Error Rate is calculated by the difference in the Uptime and Downtime Stamps. The Broker designs and decides the acceptance of the service based on Error Rate (MTBF, MTTF, MTTR) and Fault Tolerance. As, the Error Rate and Service Reliability are inversely proportional, the server with low error rate provides high reliability. Our Experimental Results with their groupings prove that the reliability can be evaluated using the Web Log File analysis.
Keywords: Composite Web Services, Error log, HTTP Status Error, Reliability, Web error.