Fault-tolerant real-time objects

Towards Real-Time Fault-Tolerant CORBA Middleware

“…Schemes like [36,37,38] have been developed and deployed that address the conflict between real-time and faulttolerance capabilities of non-CORBA systems. We propose and evaluate a strategy in Section 4 for CORBA systems that provides bounded fault-tolerance capabilities, while also maintaining the real-time properties essential to develop dependable DRE systems.…”

“…Other related work [37,38,53] presents principles and mechanisms to achieve fault-tolerance in real-time systems via Time-triggered Message-triggered Objects [54], which have been explored at the University of California, Irvine. The work that has been presented in this paper is in the context of CORBA and RT-CORBA and we expect this work to resolve some open R&D challenges [55].…”

An Approach for Adaptive Fault Tolerance in Object-Oriented Open Distributed Systems

“…The Distributed Recovery Block (DRB) scheme [12] is a tasklevel fault-tolerance scheme and executes different versions of the application task in different hardware nodes such that an occurrence of software or hardware failures in one node can be tolerated by using the result of other healthy nodes. The DRB scheme was extended in [13] to provide object-level real-time fault-tolerance (named primary shadow RTO.k replication -PSRR). We integrated ADRB and PSRR in our fault-tolerant execution modes which is shown in Figure 4 As depicted in Figure 4, the supported execution modes include the following three fault-tolerant execution modes:…”

A system for parallel media processing

“…The resource manager is responsible for evaluating the graphs presented to it to produce output data. Although the nature of the data being processed introduces real-time constraints, we reject as unecessarily limited systems which guarantee performance through static scheduling [11]. The method of evaluation selected, eduction, heavily inuences the inherent fault-tolerance of the system.…”