Pascale Minet

In this paper, we study the problem of maintaining the consistency of distributed objects in real-time. Sporadic transactions sent by a set of clients must be processed by a set servers with a bounded end-to-end response time. We propose in this article a solution, following the state machine approach, based on a particular non-preemptive scheduling denoted EDF* (Earliest Deadline First

Uniform reliable multicast protocols with a guaranteed response time are of concern in numerous distributed real-time systems (e.g. distributed transactional systems, high available systems). We focus on uniform reliable multicast protocols in processors groups. The source of a message can belong or not to the destination group. A bounded number of processors crashes and network omissions is tolerated. The uniform

We focus on non-preemptive Fixed Priority (fp) scheduling. Unlike the classical approach, where flows sharing the same priority are assumed to be scheduled arbitrarily, we assume that these flows are scheduled Earliest Deadline First (edf), by considering their absolute deadline on their first visited node. The resulting scheduling is called fp/edf *. In this paper, we establish new results for fp/edf * in a distributed context, first when flows follow the same sequence of nodes (the same path). We then extend these results when flows follow different paths. We show how to compute an upper bound on the end-to-end response time of any flow when the packet priority is computed on the first node and left unchanged on any subsequent node. This alleviates the packet processing in core nodes. For that purpose, we use a worst case analysis based on the trajectory approach, that is less pessimistic than classical approaches. We compare our results with those provided by the holistic approach: the benefit can be very high.

In this paper, we are interested in real-time flows requiring quantitative and deterministic Quality of Service (QoS) guarantees. We focus more particularly on two QoS parameters: the worst case end-to-end response time and jitter. We consider a non-preemptive scheduling of flows, called FP/FIFO, based on fixed priorities. On each node, packets are scheduled according to their fixed priority, if several

In this paper, we are interested in real-time flows requiring quantitative and deterministic Quality of Service (QoS) guarantees. We focus more particularly on two QoS parameters: the worst case end-to-end response time and jitter. We consider a non-preemptive scheduling of flows, called fp/dp \sc *^{\rm {\sc *}}, combining fixed priority and dynamic priority established on the first node visited in

We propose to use Fibre Channel (FC) technology in multimedia systems offering Video on Demand (VoD) services. The Storage Area Network (SAN) is based on (i) FC-loops connecting magnetic disks and on (ii) FC-switches connecting loops to servers. We show how to dimension FC-loops to offer a deterministic guarantee of Quality of Service to the VoD clients. The performance results

An atomic broadcast protocol should meet the three properties of unanimity, order and termination. The protocol presented here assumes weak fail-silent Network Attachment Controllers (NACs) and a broadcast channel subject to omission failures. In a failure-free environment it consists of a single phase. The response time achieved is almost the same as for a simple broadcast. In case of receivers failures, the number of phases depends on failure occurrences. In case of a broadcaster failure all the correct receivers detect the failure and all the receivers ignore the broadcast. This protocol ensures that each correct NACs has the same view of the correct NACs' list.

... (wj~)s) (1pab) s + Ark (I-(Brk)*(s)) S - A+ + A+ (B+) (s) ... B. End-to-end Response Time Distribution Let Srck be the random variable the end to end response time for a packet of class k with priority r and Srk(s) the LT of its probability density function. We have: ...

... A. Qayyum, Analysis and evaluation of channel access schemes and routing pro-tocols for wireless ... P. Jacquet, A. Laouiti, P. Minet, L. Viennot, “Performance analysis of OLSR multipoint relay ... A. Boukerche et al., “Analysis of randomized congestion control with DSDV rout-ing in ...

In this paper, we focus on the Bluetooth wireless network, analyzing its ability to support Quality of Service (QoS) requirements defined by the application. We focus on two QoS parameters : (i) an application constraint denoting the importance degree of a message, and (ii) an end-to-end delivery deadline. In a first step, we introduce a local scheduling accounting for the

The IEEE 802.15.4e MAC amendment has been proposed to meet the requirements of industrial applications. Using slotted medium access with channel hopping, the MAC layer orchestrates the medium accesses of nodes according to a given schedule. Nevertheless, this amendment does not specify how this schedule is computed. The purpose of this paper is to propose a distributed joint time slot and channel assignment, calledWave for data gathering in low power lossy networks. This schedule targets minimized data convergecast delays by reducing the total number of slots in the schedule. Moreover, Wave ensures the absence of conflicting transmissions in the schedule provided. In such a schedule, a node is awake only during its transmission slots and those of its children in the convergecast routing graph. Thus, energy efficiency is ensured. In this paper, we present Wave with its properties (e.g. support of heterogeneous traffic, support of a sink equipped with multiple interfaces, worst case delays and buffer size) and compare it with a centralized scheduling algorithm like TMCP and a distributed one like DeTAS. Simulation results show the good performance of Wave compared with TMCP. Since in an industrial environment, several routing graphs can coexist, we study how Wave supports this coexistence.

ABSTRACT In this report, we study the issue of delay optimization and energy efficiency in grid wireless sensor networks (WSNs). We focus on STDMA scheduling, where a predefined cycle is repeated, and where each node has fixed transmission opportunities during specific slots (defined by colors). We assume a STDMA algorithm that takes advantage of the regularity of grid topology to also provide a spatially periodic coloring ("tiling" of the same color pattern). In this setting, the key challenges are: 1) minimizing the average routing delay by ordering the slots in the cycle 2) being energy efficient. Our work follows two directions: first, the baseline performance is evaluated when nothing specific is done and the colors are randomly ordered in the STDMA cycle. Then, we propose a solution, ORCHID that deliberately constructs an efficient STDMA schedule. It proceeds in two steps. In the first step, ORCHID starts form a colored grid and builds a hierarchical routing based on these colors. In the second step, ORCHID builds a color ordering, by considering jointly both routing and scheduling so as to ensure that any node will reach a sink in a single STDMA cycle. We study the performance of these solutions by means of simulations and modeling. We study the performance of these solutions by means of simulations and modeling. Results show the excellent performance of ORCHID in terms of delays and energy compared to a shortest path routing that uses the delay as a heuristic.