Effective erasure codes for reliable computer communication protocols
ACM SIGCOMM Computer Communication Review
MPI code encapsulating using parallel CORBA object
Cluster Computing
Load Balancing for a Distributed CORBA-Based SCP
IS&N '00 Proceedings of the 7th International Conference on Intelligence and Services in Networks: Telecommunications and IT Convergence Towards Service E-volution
PACO++: A Parallel Object Model for High Performance Distributed Systems
HICSS '04 Proceedings of the Proceedings of the 37th Annual Hawaii International Conference on System Sciences (HICSS'04) - Track 9 - Volume 9
SPAND: shared passive network performance discovery
USITS'97 Proceedings of the USENIX Symposium on Internet Technologies and Systems on USENIX Symposium on Internet Technologies and Systems
| Hi-index | 0.00 |
Internet and Intranet repositories are store houses for multi-gigabit movie files and large multimedia documents. It thus necessitates the development of scalable architectures to locate these resources and to distribute the downloading efficiently among them. Traditionally the documents are downloaded entirely from a single server by establishing parallel connections between the clients and servers. The proposed method of parallel access downloading, i.e. paraloading, is a process that downloads parts of a document in parallel from different mirrored Common Object Request Broker Architecture (CORBA) servers. It considers the load of the servers and their distance from the clients. It is also tolerant to server failures and hence it is highly suitable for a grid-based environment. The proposed method is implemented as a service in Zen, an open source, embedded CORBA implementation based on real time CORBA, as well as real time Java specifications. Test results show that the proposed method consistently downloads files faster then the traditional method.