Presents a flexible, CORBA-compliant middle-tier server architecture which is capable of adding dependability (i.e. reliability, availability and performability) to an existing service. The architecture provides a flexible and cost-effective framework for building fault-tolerant applications via straightforward integration of legacy software. In a typical scenario, the service would be originally provided by some legacy server, which is integrated in the new system with no changes being made to it. This has two main advantages. First, the development effort is minimized, since full reuse of existing software is achieved. Second, backward compatibility is preserved since it is possible to integrate new clients with existing applications and databases, protecting the investment in legacy systems. The architecture we propose is novel, although it builds upon redundancy techniques which have already been implemented in a massive number of designs for traditional fault-tolerant applications
A CORBA-based Architecture for Adding Dependability to Legacy Servers / Cotroneo, Domenico; L., Romano; Russo, Stefano; Mazzocca, Nicola. - STAMPA. - (2000), pp. 76-82. (Intervento presentato al convegno Symposium on Software Engineering for Parallel and Distributed Systems tenutosi a Limerick, Irlanda nel maggio 2000) [10.1109/PDSE.2000.847852].
A CORBA-based Architecture for Adding Dependability to Legacy Servers
COTRONEO, DOMENICO;RUSSO, STEFANO;MAZZOCCA, NICOLA
2000
Abstract
Presents a flexible, CORBA-compliant middle-tier server architecture which is capable of adding dependability (i.e. reliability, availability and performability) to an existing service. The architecture provides a flexible and cost-effective framework for building fault-tolerant applications via straightforward integration of legacy software. In a typical scenario, the service would be originally provided by some legacy server, which is integrated in the new system with no changes being made to it. This has two main advantages. First, the development effort is minimized, since full reuse of existing software is achieved. Second, backward compatibility is preserved since it is possible to integrate new clients with existing applications and databases, protecting the investment in legacy systems. The architecture we propose is novel, although it builds upon redundancy techniques which have already been implemented in a massive number of designs for traditional fault-tolerant applicationsI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
