In a distributed computing system, a number of program modules may need to be allocated to different processors such that the reliability of executing successfully these modules is maximized and the constraints with l...
详细信息
In a distributed computing system, a number of program modules may need to be allocated to different processors such that the reliability of executing successfully these modules is maximized and the constraints with limited resources are satisfied. The problem of finding an optimal task allocation with maximum system reliability has been shown to be NP-hard;thus, existing approaches to finding exact solutions are limited to the use in problems of small size. This paper presents a hybrid particle swarm optimization (HPSO) algorithm for finding the near-optimal task allocation within reasonable time. The experimental results show that the HPSO is robust against different problem size, task interaction density, and network topology. The proposed method is also more effective and efficient than a genetic algorithm for the test-cases studied. The convergence and the worst-case characteristics of the HPSO are addressed using both theoretical and empirical analysis. (c) 2006 Elsevier Inc. All rights reserved.
We consider the automatic generation of parallel templates as basis for the implementation of n(2)-problem solvers on parallelsystems. The automatic template generator relies on the embedding of a virtual systolic lo...
详细信息
ISBN:
(纸本)0818680431
We consider the automatic generation of parallel templates as basis for the implementation of n(2)-problem solvers on parallelsystems. The automatic template generator relies on the embedding of a virtual systolic loop on the given machine's network and optimizes hyper-systolic data communication by means of Simulated Annealing techniques. We have generated a parallel template to efficiently program an n-body code on the parallel SIMD system Quadrics.
Information systems are basically distributed while object oriented brings new issues in softwareengineering. Pairing distributedsystems with object paradigms results in a distributed object model. Choosing the appr...
详细信息
ISBN:
(纸本)0780382927
Information systems are basically distributed while object oriented brings new issues in softwareengineering. Pairing distributedsystems with object paradigms results in a distributed object model. Choosing the appropriate distributed object model is a multi-criteria decision problem. Performance is a key criterion that makes software developers be able to select the object-oriented system that best fit their requirements. Classical techniques and methodologies of performance analysis are either unsuitable or unnatural to capture performance behavior of Object-Oriented (OO) systems. In this paper, we present Performance-Based Model for distributed Object-Oriented software (DOOS) and a methodology to analyze and evaluate its performance. The new model evaluates the overall time cost of DOOS system considering the communication overheads. while preserving the OO features such as encapsulations, information hiding, inheritance, Etc.
Inverse problems arise in various areas of science and engineering. These problems are not only difficult to solve numerically, but they also require a large amount of computer resources both in time and memory. It is...
详细信息
ISBN:
(纸本)9781479984480
Inverse problems arise in various areas of science and engineering. These problems are not only difficult to solve numerically, but they also require a large amount of computer resources both in time and memory. It is therefore not surprising that inverse problems are often solved using techniques from high-performance computing. We consider the parallelization of an inverse problem in the field of geothermal reservoir engineering. In this particular scientific application, the underlying software package is already parallelized using the shared-memory programming paradigm OpenMP. Here, we present an extension of this parallelization to distributed memory enabling a hybrid OpenMP/MPI parallelization. The situation is different from the standard way of hybrid parallel programming because the data structures of the OpenMP-parallelized code differ from those in the serial implementation. We exploit this transformation of the data structures in our distributed-memory strategy for parallelizing an ensemble Kalman filter, a particular method for the solution of inverse problems. We describe this novel parallelization strategy, introduce a performance model, and present timing results on a compute cluster using nodes with 2 sockets, each equipped with Intel Xeon X5675 Westmere EP processors with 6 cores. All timing results are obtained with a pure MPI parallelization without using any OpenMP threads.
Embedded designers now have the capability of offloading software routines into custom applicationspecific hardware blocks. This paper evaluates a domain-specific design system for configurable computing platforms tha...
详细信息
ISBN:
(纸本)0769523129
Embedded designers now have the capability of offloading software routines into custom applicationspecific hardware blocks. This paper evaluates a domain-specific design system for configurable computing platforms that combine processors and configurable fabrics. Although the proposed work uses floating-point and communication primitives as the specific domain tasks evaluated, the reconfigurable computing platform and the design challenges addressed will become increasingly common in a number of embedded system environments. This paper explores the added cost of hardware resources, area, and power of moving software library routines into hardware blocks in a configurable embedded system based on the MicroBlaze soft processor.
Processor scheduling in distributed-memory systems has received considerable attention in recent years. Several commercial distributed-memory systems use space-sharing processor scheduling. in space-sharing, the set o...
详细信息
ISBN:
(纸本)0818684038
Processor scheduling in distributed-memory systems has received considerable attention in recent years. Several commercial distributed-memory systems use space-sharing processor scheduling. in space-sharing, the set of processors in a system is partitioned and each partition is assigned for the exclusive use of a job. Space-sharing policies can be divided into fixed, static, or dynamic categories. For distributed-memory systems, dynamic policies incur high overhead. Thus, static policies are considered as these policies provide a better performance than the fixed policies. Several static policies have been proposed in the literature. In a previously proposed adaptive static policy, the partition size is a function of the number of queued jobs. This policy, however, tends to underutilize the system resources. To improve the performance of this policy, we propose a new policy in which the partition size is a function of the total number of jobs in the system, as opposed to only the queued jobs. The results presented here demonstrate that the new policy performs substantially better than the original policy for the various workload and system parameters. Another major contribution is the evaluation of the performance sensitivity to job structure, variances in inter-arrival times and job service times, and network topology.
Xgrid Technical Preview 2 is a distributed computing software technology from Apple Computer, Inc. Xgrid leverages the UNIX-based capabilities of Mac OS X to enable the rapid aggregation of Macintosh systems into a si...
详细信息
ISBN:
(纸本)0780321754
Xgrid Technical Preview 2 is a distributed computing software technology from Apple Computer, Inc. Xgrid leverages the UNIX-based capabilities of Mac OS X to enable the rapid aggregation of Macintosh systems into a simple yet powerful computational grid which can run a wide range of standard and custom solutions with minimal code changes. To demonstrate the technology, the software is run on a networked rack of Xserve G5 servers and a PowerBook G4 laptop. The demonstration shows how Xgrid is utilized to distribute long-running batch and parallel jobs to a local grid of Mac OS X-based computers.
This paper presents a theoretical and practical framework for constructing and reasoning about mobile agents. The framework is formulated as a process calculus and has two contributions. One of the contributions can m...
详细信息
This paper presents a theoretical and practical framework for constructing and reasoning about mobile agents. The framework is formulated as a process calculus and has two contributions. One of the contributions can model not only individual mobile agents but also a group of mobile agents because the calculus allows more than one mobile agent to be dynamically organized into a single mobile agent. The other contribution can exactly model many features of actual mobile agents, such as mobility and marshaling, which are often ignored in other existing frameworks but may seriously affect the correctness of mobile agents. To demonstrate the utility of the calculus, we constructed a practical mobile agent system whose agents can be naturally and strictly specified and verified in the calculus. The system also offers a security mechanism for mobile agents by using well-defined properties of the calculus.
Rejuvenation is a technique expected to mitigate failures in HPC systems by replacing, repairing, or resetting system components. Because of the small overhead required by software rejuvenation, we primarily focus on ...
详细信息
OpenMP has emerged as an important model and language extension for shared-memory parallel programming. On shared-memory platforms, OpenMP offers an intuitive, incremental approach to parallel programming. In this pap...
详细信息
暂无评论