In this paper, we have presented a reliable data transfer mechanism using directed diffusion in WSNs (Wireless Sensor networks). This mechanism involves selecting a path with higher reachability and transferring data ...
详细信息
ISBN:
(纸本)9780889866379
In this paper, we have presented a reliable data transfer mechanism using directed diffusion in WSNs (Wireless Sensor networks). This mechanism involves selecting a path with higher reachability and transferring data along the path chosen, which is based on the end-to-end reliability calculated by the dissemination procedure of Interest packets, while each node of a sensor network maintains only the information on its neighborhood. We performed various experiments changing the link error rates and the number of nodes and discovered that this proposed mechanism improves event-to-sink data transfer reliability in WSNs. We also found that this mechanism spreads evenly traffic load over the network and reduces energy consumption, which in turn prolongs network lifetime.
Large-scale interconnection networks require fault-tolerant features as well as high performance. Since a large-scale network consists of huge number of independent routers, effective globally-optimum algorithms are r...
详细信息
ISBN:
(纸本)9780889866379
Large-scale interconnection networks require fault-tolerant features as well as high performance. Since a large-scale network consists of huge number of independent routers, effective globally-optimum algorithms are required. We introduce a quasi-global control mechanism to solve the problem and propose Fault-Tolerant Cross-Line as a novel selection function of routing algorithm. The proposed method makes use of fault location information as well as congestion information so that it can properly control under both faulty and congested situations. Preliminary evaluation results show essential effectiveness of the proposed method.
In this paper, we propose and analyse a static priority scheduling scheme providing class-based quality of service (QoS) with channel error compensation. Our method can be used in the wireless IP networks. With this n...
详细信息
ISBN:
(纸本)088986568X
In this paper, we propose and analyse a static priority scheduling scheme providing class-based quality of service (QoS) with channel error compensation. Our method can be used in the wireless IP networks. With this new method, we are able to successfully employ good compensation of channel errors for flow admission. Our method is scalable to large systems. The detailed algorithm and analysis show that our method can improve some classes, such as CBR flow and the real-time VRB flow, and at the same time maintain low complexity.
An Explicit Data Organization (EDO-SIMD) instruction set architecture (ISA) is proposed in this paper to reduce the data organization overhead of SIMD (Single Instruction Multiple Data) for media processors. It explic...
详细信息
ISBN:
(纸本)9780889866379
An Explicit Data Organization (EDO-SIMD) instruction set architecture (ISA) is proposed in this paper to reduce the data organization overhead of SIMD (Single Instruction Multiple Data) for media processors. It explicitly describes the data organization information in instruction words and merges the data organization with computation and store operation. An implementation of EDO-SIMD ISA based on a baseline SIMD processor is described and cycle accurate simulator for evaluation is designed. Simulation results show that, relative to the baseline SIMD architecture, EDO-SIMD ISA can achieve 1.34 to 1.40 speedups for the benchmark of real time H.264/AVC decoder and reduce 17.7% of the code size with only 0.49% increase in hardware area.
A computational Grid, as an emerging framework for providing globally distributedcomputing resources to our desktop, has been established as a new programming environment. In particular, a new Grid research for utili...
详细信息
ISBN:
(纸本)9780889866379
A computational Grid, as an emerging framework for providing globally distributedcomputing resources to our desktop, has been established as a new programming environment. In particular, a new Grid research for utilizing a large number of busy computers is indispensable for further development of Grid. In this paper, we concentrate on Grid scheduling problem. To cope with unstable availability, low reliability, and mixed operation policies that are still serious problems in metascheduling, we propose ASF, Agent-based Scheduling Framework whose idea is dual to the conventional metascheduling. ASF is composed of a single discreet metascheduler and a collection of autonomous agents attached to each computing resource manager. Each agent autonomously finds jobs to be processed, instead of being assigned a job to by a omnipotent metascheduler. We implement a prototype of ASF, and prove its effectiveness. Our experiment shows that the total elapsed time of job processing is reduced by 11%.
This paper presents a proposal of ad hoc network based on TETRA technology. The ad hoc deployment without the need of infrastructure makes Wireless Mesh networks (WMNs) a very suitable solution for emergency communica...
详细信息
ISBN:
(纸本)9780889866386
This paper presents a proposal of ad hoc network based on TETRA technology. The ad hoc deployment without the need of infrastructure makes Wireless Mesh networks (WMNs) a very suitable solution for emergency communications systems. On the other hand, TETRA radio spectrum is more suitable than that of IEEE 802.11 standard in which most of the WMNs are currently based. Making use of the technology that TETRA offers we propose a multi-hop ad hoc network which will be part of a TETRA mesh network solution.
Micro sensor nodes are now easily available and are very cheap. These resource constrained nodes are spread in remote locations to gather data. There are some energy efficient clustering protocols that are known to wo...
详细信息
ISBN:
(纸本)9780889866386
Micro sensor nodes are now easily available and are very cheap. These resource constrained nodes are spread in remote locations to gather data. There are some energy efficient clustering protocols that are known to work well for sensor networks. The analyses of these algorithms that have been done in the past are based on the energy consumption only. In this paper, we develop a performance measure for these algorithms that includes the energy spent, the time delay incurred in the whole process and the information loss. We have formulated a novel metric, the information loss metric, which to the best of our knowledge has not been previously addressed. We then optimize the joint performance measure which is a combination of these different metrics to yield the optimal clustering configuration.
In this study we evaluate and compare the performance of our load balancing technique on irregular P2P systems embedded in two regular topologies: 1) Hypercube and 2) TreeP. Hypercube is one of the most studied interc...
详细信息
ISBN:
(纸本)9780889866379
In this study we evaluate and compare the performance of our load balancing technique on irregular P2P systems embedded in two regular topologies: 1) Hypercube and 2) TreeP. Hypercube is one of the most studied interconnection topologies and exhibits powerful interconnection features, while TreeP is Tree-based P2P network architecture and is based on a tessellation of a 1-D space. The load balancing technique employed is a two step strategy. In the first phase, it maps any irregular network topology to a regular one. In the second phase the load is balanced among the nodes using PSLB algorithm. In this paper we study and compare the performance of this two-step strategy on hypercube and TreeP topologies. This strategy is proven to be efficient and does not introduce a considerable overhead as shown in the experimental results.
Operating system virtualization has recently become a popular technique to achieve better resource utilization in so-called '' server farm '' environments. This technique provides a virtual hardware in...
详细信息
ISBN:
(纸本)9780889866379
Operating system virtualization has recently become a popular technique to achieve better resource utilization in so-called '' server farm '' environments. This technique provides a virtual hardware interface on top of which one can run multiple instances of popular operating systems. The Xen Virtual Machine Monitor is an implementation of operating system virtualization that supports live migration, the transfer of a virtual operating system from one physical machine to another with minimal down time. We have utilized this capability to implement a monitoring and dynamic reconfiguration daemon that attempts to equalize the load on all host nodes in a group of machines running Xen. We have also implemented a simulator for testing balancing algorithms. Experiments using these tools have provided insight into the redistribution of virtualized operating systems and how this differs from the more thoroughly-studied problem of process-level load balancing.
Scheduling tasks onto the processors of a parallel system is a crucial part of program parallelisation. Due to the NP-hardness of the task scheduling problem, scheduling algorithms are based on heuristics that try to ...
详细信息
ISBN:
(纸本)9780889866379
Scheduling tasks onto the processors of a parallel system is a crucial part of program parallelisation. Due to the NP-hardness of the task scheduling problem, scheduling algorithms are based on heuristics that try to produce good rather than optimal schedules. Nevertheless, in certain situations it is desirable to have optimal schedules, for example for time critical systems or to evaluate scheduling heuristics. This paper proposes a scheduling algorithm based on A* that can produce optimal schedules in reasonable time for small task graphs. A* is a best-first state space search algorithm. In comparison to a previous approach, the here presented scheduling algorithm has a significantly reduced search space due to a much improved cost function f (s) and additional pruning techniques. Experimental results reveal the relation between the runtime of the algorithm and the structure of the task graphs. Further, it is shown that the proposed algorithm significantly outperforms the previous approach.
暂无评论