The essential distributed objects survival guide
The essential distributed objects survival guide
Concurrent Systems: An Integrated Approach to Operating Systems, Database, and Distributed ...
Concurrent Systems: An Integrated Approach to Operating Systems, Database, and Distributed ...
Distributed Systems: Concepts and Design
Distributed Systems: Concepts and Design
CORBA Based Runtime Support for Load Distribution and Fault Tolerance
IPDPS '00 Proceedings of the 15 IPDPS 2000 Workshops on Parallel and Distributed Processing
Distributed computing: moving from CGI to CORBA
ATEC '00 Proceedings of the annual conference on USENIX Annual Technical Conference
Hi-index | 0.00 |
As more and more computers and workstations enter the workplace they are inevitably connected to a network. Networks provide the interconnection necessary for computers to share common data, peripherals and other system resources. Distributed computing allows network applications to access functions or processes on remote computers. Developing network applications specifically to interact and draw upon resources of multiple computers creates the groundwork for a distributed system or distributed computing environment (DCE). An ideal distributed system is self monitoring and resilient to failures. In the event of a failure the system should dynamically reconfigure itself with automatic fail-over for applications that fall victim to the fault. Transparency of fault tolerant mechanisms is desirable, especially when introducing legacy applications into the distributed system. The reduction of application development efforts heavily relies on the availability of portable, non-invasive, fault tolerance providing extensions, which introduce mechanisms for uninterruptible service by insertion into existing distributed applications. To test the potential for addressing some of these desired capabilities for a distributed system implemented within a CORBA distributed computing environment, the Interactive-Group Object-Replication (IGOR) system was developed. IGOR is a system of objects that provides fault tolerance through object replication by arranging replicas in fault tolerant groups which interact to provide access to redundant data and services. For purposes of portability, interoperability and to evaluate the CORBA environment, IGOR was designed with the constraint of lying entirely within the CORBA architecture and using IIOP as the communication protocol. This guarantees its portability over changes in platform and network technologies. The IGOR system is re-configurable and its fault tolerance mechanisms are completely transparent to client applications.