A case for redundant arrays of inexpensive disks (RAID)
SIGMOD '88 Proceedings of the 1988 ACM SIGMOD international conference on Management of data
A scalable HTTP server: the NCSA prototype
Selected papers of the first conference on World-Wide Web
Performance and scalability of EJB applications
OOPSLA '02 Proceedings of the 17th ACM SIGPLAN conference on Object-oriented programming, systems, languages, and applications
Performability evaluation: where it is and what lies ahead
IPDS '95 Proceedings of the International Computer Performance and Dependability Symposium on Computer Performance and Dependability Symposium
State Maintenance and its Impact on the Performability of Multi-tiered Internet Services
SRDS '04 Proceedings of the 23rd IEEE International Symposium on Reliable Distributed Systems
C-JDBC: flexible database clustering middleware
ATEC '04 Proceedings of the annual conference on USENIX Annual Technical Conference
Session state: beyond soft state
NSDI'04 Proceedings of the 1st conference on Symposium on Networked Systems Design and Implementation - Volume 1
Timeliness and transaction management in extended enterprises
International Journal of Business Information Systems
Markovian workload modeling for Enterprise Application Servers
C3S2E '09 Proceedings of the 2nd Canadian Conference on Computer Science and Software Engineering
| Hi-index | 0.00 |
With the development of Internet-based business, Web applications are becoming increasingly complex. The J2EE specification aims at enabling the design of such web application servers. These servers have to ensure scalability and availability of the supported applications. Scalibility can be achieved using replication techniques or partitionning techniques. The aim of this paper is to compare these approaches. In a J2EE web application server, one important component is the EJB tier. In this context, the JOnAS web application server provides an example of EJB replication system called CMI (Cluster Method Invocation). In a first step, this paper presents a performance evaluation of CMI. It then introduces incrementally an alternative scheme based on partitionning and shows the performance benefits compared to CMI.