Search

International audience; During interactions, system actors may face up to misunderstandings when their local visions contain inconsistent data about a same fact. Misunderstandings in interaction are likely to reduce interactivity performances (deviation or deadlock) or even affect overall system behavior. In this paper, we present agent-based architecture and scenario-structuring approach to deal with such misunderstandings and consistency. It is based on the notion of "situation" that is an elementary building block dividing the interactions between actors into contextual scenes. This model not only supports the scenario execution, but the consistency management as well. In order to organize and control the interactions, a "situation" contextualizes system's actors' interaction and activity, and includes prevention and tolerances mechanisms to deal with the misunderstandings and their causes. We also simulation experimentation on an Online Distance Learning case study.

We advocate the role of tuple-based coordination languages as effective tools for event-driven programming of situated multi-agent systems (MAS). By focussing on logic-based coordination artefacts, we discuss the benefits of exploiting ReSpecT tuple centres as event-driven abstractions for MAS coordination.

International audience; NFSG aims at providing a solution for file accesses within a cluster of clusters. Criteria of easiness (installation, administration, usage) but also efficiency as well as a minimal hardware and software intrusivity have led our developments. By using several facilities such as distributed file systems (NFSP) and a high-performance data transfer utility (GXfer), we hope to offer a software architecture fully compatible with the ubiquitous NFS protocol. Thanks to a distributed storage (especially multiple I/O servers provided by NFSP), several parallel streams may be used when copying a file from one cluster to another within a same grid. This technique improves data transfers by connecting distributed file system at both ends. The GXfer component implements this functionality. Thus, performances only reachable with dedicated and expensive hardware may be achieved.

International audience; In this paper we consider the use of a supercomputer with a hardware shared memory versus a cluster of workstations using a software Distributed Shared Mem-ory (DSM). We focus on ray tracing applications to compare both architectures. We have ported Stingray, a parallel cone tracer developed on a SGI Origin 2000 super-computer, on a cluster using a Scalable Coherent Interface (SCI) network and a software DSM called SciFS. We present concepts of cone tracing with Stingray, concepts of SCI cluster with a DSM and the implementa-tion issues. We compare the results obtained with the two architectures and we discuss the trade-off - price/performance/programming ease - of both architectures. We show with Stingray that a modest 12 nodes SCI cluster with an efficient software DSM is 5 times cheaper and can perform up to 2.3 times better than a SGI Origin 2000 with 6 processors. We think that a software DSM is well suited for this kind of applications and provides both ease of programming and scalable per-formance.

Non-Local Means (NLM) algorithm is widely considered as a state-of-the-art denoising filter in many research fields. High computational complexity led to implementations on Graphic Processor Unit (GPU) architectures, which achieve reasonable running times by filtering, slice-by-slice, 3D datasets with a 2D NLM approach. Here we present a fully 3D NLM implementation on a multi-GPU architecture and suggest its high scalability. The performance results we discuss encourage the coding of further filter improvements and the investigation of a large spectrum of applicative scenarios. © 2013 Polish Information Processing Society.

In this paper we describe a parallel implicit method based on radial basis functions (RBF) for surface reconstruction. The applicability of RBF methods is hindered by its computational demand, that requires the solution of linear systems of size equal to the number of data points. Our reconstruction implementation relies on parallel scientific libraries and is supported for massively multi-core architectures, namely Graphic Processor Units (GPUs). The performance of the proposed method in terms of accuracy of the reconstruction and computing time shows that the RBF interpolant can be very effective for such problem., arXiv admin note: text overlap with arXiv:0909.5413 by other authors

This paper grounds on the SAPERE project (Self-Aware PERvasive Service Ecosystems), which aims at proposing a multi-agent framework for pervasive computing, based on the idea of making each agent (service, device, human) manifest its existence in the ecosystem by a Live Semantic Annotation (LSA), and of coordinating agent activities by a small and fixed set of so-called eco-laws - sort of chemical-like reactions over patterns of LSAs. System dynamics in SAPERE is complex because of opennes and due to the self-* requirements imposed by the pervasive computing setting: a simulation framework is hence needed for what-if analysis prior to deployment. In this paper we present a prototype simulator we are developing. Due to the role of chemical-like dynamics, this is based on a variation of an existing SSA (Stochastic Simulation Algorithm), suitable tailored to the specific features of SAPERE, including dynamicity of network topology, pattern-based application of eco-laws, and temporal triggers. The simulator is tested on a crowd steering scenario where the navigation of groups is guided towards the preferential destination by showing the direction on public or private screens.