Partial and dynamic reconfiguration provides a relevant new dimension to design efficient parallel embedded systems. However, due to the encasing complexity of such systems, ensuring the consistency and parallelism ma...
详细信息
ISBN:
(纸本)9781467325721
Partial and dynamic reconfiguration provides a relevant new dimension to design efficient parallel embedded systems. However, due to the encasing complexity of such systems, ensuring the consistency and parallelism management at runtime is still a key challenge. So architecture models and design methodology are required to allow for efficient component reuse and hardware reconfiguration management. this paper presents a distributed persistence management model and its implementation for reconfigurable multiprocessor systems on dynamically reconfigurable circuits. the proposed approach is inspired from the well-known component based models used in software applications development. Our model is based on membranes wrapping the systems components. the objective is to improve design productivity and ensure consistency by managing context switching and storage using modular distributed hardware controllers. these membranes are distributed and optimized withthe aim to design self-adaptive systems by allowing dynamic changes in parallelism degree and contexts migration. simulation and synthesis results are given to show performances and effectiveness of our methodology.
Motion planning is a fundamental problem in a number of application areas, including robotics, automation, and virtual reality! this paper describes a parallel implementation of a motion planning algorithm particularl...
详细信息
ISBN:
(纸本)0769500595
Motion planning is a fundamental problem in a number of application areas, including robotics, automation, and virtual reality! this paper describes a parallel implementation of a motion planning algorithm particularly suited for complex systems characterized by many degrees of freedom. nle implementation is based on the concurrent exploration of the search space by a randomized planner replicated on each node of the parallel architecture. All processing elements compete to obtain a solution over the entire search space in an OR-parallel fashion, Reported results refer to a low-cost cluster of PCs and an SGI Onyx2 parallel machine. the experiments emphasize the effectiveness of the approach for complex, high-dimensionality planning problems.. We believe that the approach may, be useful in other complex search problems, especially when the parallel architecture exhibits relatively high communication latency.
the bubble algorithm evaluated in this paper assures message deadlock freedom in k-ary n-cube network without wing virtual channels. this algorithm is based both on a dimension order routing (DOR) and on a restricted ...
详细信息
ISBN:
(纸本)0769500595
the bubble algorithm evaluated in this paper assures message deadlock freedom in k-ary n-cube network without wing virtual channels. this algorithm is based both on a dimension order routing (DOR) and on a restricted injection policy extended to the dimension changes. An exhaustive comparison between the bubble mechanism and the classical deterministic virtual channels solution is presented here. For that purpose, the message router of both proposals has been designed by using VHDL descriptions and the Synopsys VLSI CAD tool. Additionally :formal models of the routers, based on colored Petri nets, have been carried out together withsimulation techniques in order to assure the validation of the results and shorten the design cycle. the performance evaluation of n-dimension tori highlights the benefits of the bubble algorithm as boththe temporal delay and the necessary silicon area of the message router ape reduced.
Decentralized control is composed of more than two subsystems. Subsystems communicate each other to control the whole system. In this paper, an information of mechanical constraint for parallel manipulators analyzed, ...
详细信息
ISBN:
(纸本)0780374797
Decentralized control is composed of more than two subsystems. Subsystems communicate each other to control the whole system. In this paper, an information of mechanical constraint for parallel manipulators analyzed, When some joints of multi-degrees of freedom parallel manipulator are set to be passive, excessive interference force can be reduced. A case one joint of 3-linked arms is set to be passive is treated. the influence of a passive joint to the tip of the arm is derived by the configuration and the calculated reference torque. In the active joints, the interference force information from other joints are used to compensate the effect of passive joint. Such a information is also used for switching passive joint. As a result, the avoidance of singular point is realized. Taking advantage of the parallel manipulator which has redundant drive joints, fault compensation is achieved by extending the communication of subsystems. Comparing withthe centralized control represents the availability of the decentralized control system by simulation.
In this paper, we study the load balancing problem for PCS simulation systems, and focus upon static strategies in order to reduce the synchronization overhead of SWiMNet, a parallel PCS simulation testbed devoloped a...
详细信息
ISBN:
(纸本)3540414290
In this paper, we study the load balancing problem for PCS simulation systems, and focus upon static strategies in order to reduce the synchronization overhead of SWiMNet, a parallel PCS simulation testbed devoloped at UNT. Performance results on a cluster of workstations indicate clearly the significant impact of load balancing on the efficient implementation of SWiMNet and its two stage design.
the paper describes the architecture and the design of PargreSQL parallel database management system (DBMS) for distributed memory multiprocessors. PargreSQL is based upon PostgreSQL open-source DBMS and exploits part...
详细信息
the paper describes the architecture and the design of PargreSQL parallel database management system (DBMS) for distributed memory multiprocessors. PargreSQL is based upon PostgreSQL open-source DBMS and exploits partitioned parallelism.
作者：
Boukerche, AUniv N Texas
Dept Comp Sci Parallel Simulat & Distributed Syst PARADISE Res Denton TX 76203 USA
distributed computation among multiple processors is one approach to reducing simulation time for large VLSI circuits designs. In this paper, we investigate conservative parallel discrete event simulation for logic ci...
详细信息
ISBN:
(纸本)3540414290
distributed computation among multiple processors is one approach to reducing simulation time for large VLSI circuits designs. In this paper, we investigate conservative parallel discrete event simulation for logic circuits on multiprocessor machines. the synchronization protocol makes use of the Semi-Global Time-of-Next- Event (SGTNE/TNE) suite of algorithms. Extensive simulation experiments were conducted to validate our model using several digital logic circuits on an Intel Paragon machine.
the proceedings contain 24 papers. the topics discussed include: comprehension with[in] virtual environment visualizations;improving the visibility of graphical program displays: an experimental study;reconstructing o...
ISBN:
(纸本)0769501796
the proceedings contain 24 papers. the topics discussed include: comprehension with[in] virtual environment visualizations;improving the visibility of graphical program displays: an experimental study;reconstructing ownership architectures to help understand software systems;an architecture for automated software maintenance;extending software quality assessment techniques to java systems;improving software comprehension through an automated dependency tracer;an application of fuzzy reasoning to support automated program comprehension;restructuring parallel programs by transformation of point-to-point interactions into collective communication;and comprehending a complex, distributed, object-oriented software system: a report from the field.
the fifth generation (5G) communication technologies are envisioned to provide higher data rates, excellent user-coverage, extremely low latency and power consumption. Such cellular networks will adopt a heterogeneous...
详细信息
ISBN:
(纸本)9781450350495
the fifth generation (5G) communication technologies are envisioned to provide higher data rates, excellent user-coverage, extremely low latency and power consumption. Such cellular networks will adopt a heterogeneous multi-tier architecture consisting of macrocells, device-to-device networks, and different types of small cells to serve users with required quality-of-service (QoS). this heterogeneous multi-tier architecture poses various research challenges such as resource allocation and interference management in 5G networks. the goal of our proposed distributed algorithm is to allocate the resources to the uplink transmitters which results in higher spectral efficiency and maximize the users' data rate. Our proposed scheme gives efficient (higher data rate) results than existing work, even in the complex multi-tier networks as pointed out in the simulation analysis section.
暂无评论