Abstract Advance Planning and Reservation in a Grid System allows applications to request resources from multiple scheduling systems at a specific time in future and thus gain simultaneous access to sufficient resources for their execution. Existing advance reservation strategy will reject incoming reservation if requested resources are not available at that exact time. Therefore impact of advance reservations is decreasing resource... Advance Planning and Reservation in a Grid System allows applications to request resources from multiple scheduling systems at a specific time in future and thus gain simultaneous access to sufficient resources for their execution. Existing advance reservation strategy will reject incoming reservation if requested resources are not available at that exact time. Therefore impact of advance reservations is decreasing resource utilization due to fragmentations. This paper proposes a novel advance planning and reservation strategy namely First Come First Serve Ejecting Based Dynamic Scheduling (FCFS-EDS) to increase resources utilization in a grid system for MPI jobs. MPI jobs are jobs that need more than one compute node of their execution and must start at the same time on each of the compute node. To achieve this we introduce a new notion that maps a user job to a virtual compute nodes (called logical view) which are subsequently mapped to actual compute nodes (called physical view) at the time of execution. A lemma ensures the success of such a mapping with increased resource utilization. Read More Read Less

Abstract In this paper, we propose an optimal network selection technique for vertical handover. The proposed technique allows mobile users to select the best access network with seamless manner and to reduce the processing delay. Thus we aim to exploit the minimum of criteria which should be used for network selection. Our optimal network selection technique combines two multi attribute decision... In this paper, we propose an optimal network selection technique for vertical handover. The proposed technique allows mobile users to select the best access network with seamless manner and to reduce the processing delay. Thus we aim to exploit the minimum of criteria which should be used for network selection. Our optimal network selection technique combines two multi attribute decision making (MADM) methods such as analytic network process (ANP) method to weigh the criteria, and the grey relational analysis (GRA) to rank the available networks. Simulation results and discussion are provided to illustrate the effectiveness of our optimal network selection technique Read More Read Less

Abstract Wireless Sensor Networks (WSNs) have been regarded as distinguished Ad Hoc networks that can be used for specific applications. Since a WSN consists of potentially hundreds of low cost, small size and battery powered sensor nodes, it has more potentials than other Ad Hoc networks to be deployed in many emerging applications. The WSNs raised many new challenges such as:... Wireless Sensor Networks (WSNs) have been regarded as distinguished Ad Hoc networks that can be used for specific applications. Since a WSN consists of potentially hundreds of low cost, small size and battery powered sensor nodes, it has more potentials than other Ad Hoc networks to be deployed in many emerging applications. The WSNs raised many new challenges such as: the design of embedded sensors and wireless networking technologies, in other words the routing protocols. Scalability of routing protocols used in WSNs is a critical issue due to the extremely high node numbers and relatively high node density. A good routing protocol has to be scalable and adaptive to the changes in the network topology. Thus protocols must perform well as the network grows larger or as the workload increases. Many Ad Hoc routing protocols such as Ad Hoc On demand Distance Vector (AODV), Dynamic Source Routing (DSR), and Destination Sequenced Distance Vector (DSDV), have been proved to perform well on the Mobile Ad Hoc networks (MANETs). In this paper, a performance study is carried out, using some simulation network models, to investigate how well AODV, DSR, and DSDV routing protocols work on WSNs, in static environments, using NS-2 simulator. The performance study focuses on the impact of the network size, network density (up to 500 nodes), number of sources (data connections), and the type of generated traffic model. The performance metrics to be used in this work are average end-to-end delay, packet delivery fraction (PDF), routing overheads (ROH), and average energy consumption per received packet. Read More Read Less

Abstract Safety is one of the most important issues in VANETS. It includes collision avoidance systems and alert dissemination applications. In this paper we focus on the second issue because the availability of the traffic road information in time can enlarge the conductor’s visibility. Classical solutions are based on the well known flooding strategy which uses multi-hop broadcast to reach a... Safety is one of the most important issues in VANETS. It includes collision avoidance systems and alert dissemination applications. In this paper we focus on the second issue because the availability of the traffic road information in time can enlarge the conductor’s visibility. Classical solutions are based on the well known flooding strategy which uses multi-hop broadcast to reach a large number of vehicles. Their main concern is how to decrease the number of relay nodes because this is the key to reduce broadcast storm effects. The majority of existing solutions discards the use of unicast messages. The contribution of this paper consists essentially in using both the unicast and the broadcast modes. Simulation shows that the proposed protocol achieves low latency in delivering alerts. Read More Read Less