DARX - a framework for the fault-tolerant support of agent software

Abstract

This paper presents DARX, our framework for building applications that provide adaptive fault tolerance. It relies on the fact that multi-agent platforms constitute a very strong basis for decentralized software that is both flexible and scalable, and makes the assumption that the relative importance of each agent varies during the course of the computation. DARX regroups solutions which facilitate the creation of multi-agent applications in a large-scale context. Its most important feature is adaptive replication: replication strategies are applied on a per-agent basis with respect to transient environment characteristics such as the importance of the agent for the computation, the network load or the mean time between failures.Firstly, the interwoven concerns of multi-agent systems and fault-tolerant solutions are put forward. An overview of the DARX architecture follows, as well as an evaluation of its performances. We conclude, after outlining the promising outcomes, by presenting prospective work.

Supporting component-based failover units in middleware for distributed real-time and embedded systems

“…The Time-triggered Message-triggered Objects (TMO) project [28] considers replication schemes such as the primary-shadow TMO replication (PSTR) scheme, for which recovery time bounds can be quantitatively established, and real-time fault tolerance guarantees can be provided to applications. DARX [29] provides adaptive fault-tolerance for multi-agent software platforms by dynamically changing replication styles in response to changing resource availabilities and application performance.…”

Specification of an exception handling system for a replicated agent environment

“…The concepts exposed in this section are derived from the Dimax software that combines the Dima multi-agent system [7] and the Darx fault-tolerant middleware [12]. Initial names and principles are generalized here and adapted to the agent interaction scheme we studied in our previous work [5], namely peer-topeer service exchanges.…”

“…Agents are executed on a middleware which provides a fault-tolerant execution context thanks to a replication mechanism [12]. The execution context consists of a set of distributed replication servers which manage the execution of tasks (ReplicatedTask class of Fig.…”

“…Replication handles failures whereas exceptions enable programmers to dynamically handle those situations that prevent software from running normally. An agent replication system [12,7] is able to replace an agent (provided he has been replicated) that fails by one of its (active or passive) replicas. This replacement is as seamless as possible to software users and does not require any additional code from programmers.…”

“…The bases of the study are our Sage exception handling system dedicated to agents [19], components [20] and active objects [5] and the Dimax replicated agent system [7,12] abstracted in Sect. 2.…”