Cluster of Workstations is becoming an important kind of parallel computing platform. Two issues are essential for promoting the popularization of COW: high-performance communication mechanism and powerful programming...
详细信息
ISBN:
(纸本)0818682596
Cluster of Workstations is becoming an important kind of parallel computing platform. Two issues are essential for promoting the popularization of COW: high-performance communication mechanism and powerful programming environment. In this paper Mie present our research work on COW, which aims at giving a practical solution to the above issues. In order to build a high performance communication system, we design a reduced communication protocol in addition to making use of high-speed network hardware. Based on PVM, we implement a practical programming environment which include parallel compiling systems, parallel debugger, load balancing tool and execution fault recovery tool. Some performance results are given in this paper.
This paper presents the BaLinda model, based on last in/first out threads that interact via a shared tuplespace, and discusses the idea of using function-based objects as the basic unit of parallel execution and the h...
详细信息
ISBN:
(纸本)0818682596
This paper presents the BaLinda model, based on last in/first out threads that interact via a shared tuplespace, and discusses the idea of using function-based objects as the basic unit of parallel execution and the hierarchical structure to partition tuplespaces. It is argued that the two-level parallel execution, both within. and between objects, are well suited to scalable parallel platforms with shared memory nodes connected by high speed networks.
parallelism suffers from a lack of programming languages both simple to handle and able to take advantage of the power of present parallel computers. If parallelism expression is too high level, compilers have to perf...
详细信息
ISBN:
(纸本)0818678836
parallelism suffers from a lack of programming languages both simple to handle and able to take advantage of the power of present parallel computers. If parallelism expression is too high level, compilers have to perform complex optimizations leading often to poor performances. One the other hand, too low level parallelism transfers difficulties reward the programmer. In this paper, we propose a new programming language that integrates both a synchronous data-parallel progamming model and an asynchronous execution model. The synchronous data-parallelprogramming model allows a safe program designing. The asynchronous execution model yields an efficient execution on present MIMD architectures without any program transformation. Our language relies on a logical instruction ordering exploited by specific send/receive communications. It allows to express only the effective data dependences between processors. This ability is enforced by a possible send/receive unmatching useful for irregular algorithms. A sparse vector computation exemplifies our language potentialities.
Many problems of distributed object-oriented applications can be uniformly resolved in the frame of approach based on the concept of cover. The cover is defined as an environment that transparently controls all aspect...
详细信息
ISBN:
(纸本)0818682596
Many problems of distributed object-oriented applications can be uniformly resolved in the frame of approach based on the concept of cover. The cover is defined as an environment that transparently controls all aspects of object's community life: creation, interaction etc. To enable transparency, an object-oriented application must obey a principle of late binding, a reference to server object being obtained by the client at run time from a system environment. To implement cove services, the technique of metaobject control is applied, which provides extensions of program's semantics without changing the program code, by means of attaching additional method calls to each application object invocation. A special language (TL) in which the user can incrementally define new metaservices is described and illustrated by numerous examples.
The computation of autocorrelation matrix is used heavily in several areas including signal and image processing, where parallel and application-specific architectures are also being increasingly used. Therefore, an e...
详细信息
ISBN:
(纸本)0818682596
The computation of autocorrelation matrix is used heavily in several areas including signal and image processing, where parallel and application-specific architectures are also being increasingly used. Therefore, an efficient scheme to compute autocorrelation matrix on parallelarchitectures has tremendous benefits. In this paper, a parallel algorithm for the computation of autocorrelation matrix on 2-D mesh is presented. The computation requirements for the elements of the autocorrelation matrix is highly skewed and the proposed algorithm attempts to balance the computation load, without requiring an external load balancing algorithm or processor. In this sense, the load balancing is embedded within the algorithm. The exact number of computation steps are derived. The time complexity of the proposed algorithm is shown to be within twice the optimal (or lower bound). It is also shown to have twice the speedup of a straight-forward parallel algorithm.
In this paper we consider the determination of allocation functions as a part of the design of massively parallel processor arrays for algorithms which can be represented as systems of uniform recurrence equations. Th...
详细信息
ISBN:
(纸本)0780342291
In this paper we consider the determination of allocation functions as a part of the design of massively parallel processor arrays for algorithms which can be represented as systems of uniform recurrence equations. The objective is to find allocation functions minimizing the necessary chip area for a hardware implementation of the processor array. We propose an algorithm approximately minimizing the number of processors under consideration of the necessary chip area needed to implement the processors of the processor array. The arising optimization problems can be solved using integer linear programming.
Composed of various topologies, the k-ary n-cube system is desirable for accepting and executing topologically different tasks. In this paper, we propose a new allocation strategy 20 utilize the large amount of proces...
详细信息
ISBN:
(纸本)0818682596
Composed of various topologies, the k-ary n-cube system is desirable for accepting and executing topologically different tasks. In this paper, we propose a new allocation strategy 20 utilize the large amount of processor resources in the k-ary n-cubes. Our strategy is an extension of the TC strategy on hypercubes and is able to recognize all subcubes with different topologies. Simulation results show that with such full sub-cube recognition ability and no internal fragmentation our strategy depicts constantly better performance than the other strategies, such as the Free-list strategy on k-ary n-cubes and the Sniffing strategy.
Techniques of customizing and extending operating systems (OSs) have a growing impact on system architectures in the field of distributed computing and parallelprogramming. Even if traditional methods of adaption hav...
详细信息
ISBN:
(纸本)0780342291
Techniques of customizing and extending operating systems (OSs) have a growing impact on system architectures in the field of distributed computing and parallelprogramming. Even if traditional methods of adaption have been limited to the user-level, modern OSs cannot do without kernel support. Hence concepts and structures of microkernel architectures must be redefined to meet the requirements of today's and future applications. In this paperwork we are proposing a new customizable low-level OS architecture - the Dycos kernel. In a first part we will discuss customization demands on microkernels. In a second part we are passing over to describe the basic kernel concept. Dycos is an object-based approach providing a toolbox of operations to build user-definable compositions of kernel structures. The Dycos approach has been evaluated on a Solaris 2.5.1 platform.
The thesis of this research is that the basic of exposing the parallelism in a given application should be left to the algorithm designer, who has intimate knowledge of the application characteristics. On the other ha...
详细信息
ISBN:
(纸本)0818682596
The thesis of this research is that the basic of exposing the parallelism in a given application should be left to the algorithm designer, who has intimate knowledge of the application characteristics. On the other hand, the task of limiting the parallelism in a chosen parallel algorithm is best handled by the compiler or operating system for the target MPP machine. Toward this end, we have developed CASS (for Clustering And Scheduling System), a task management system that provides facilities for automatic granularity optimization and task scheduling of parallel programs on distributed memory parallelarchitectures. Our tool environment, CASS, consists of a two-phase method of compiler-time scheduling, in which task clustering is performed prior to the actual scheduling process. The clustering module identifies the optimal number of processing nodes that the program will require to obtain maximum performance on the target parallel machine. The scheduling module maps the clusters onto a fixed number of processors and determines the order of execution of tasks in each processor.
The mpC language is an ANSI C superset supporting modular parallelprogramming for distributed memory machines. It allows the user to specify dynamically an application topology, and the mpC programming environment us...
详细信息
ISBN:
(纸本)0818678704
The mpC language is an ANSI C superset supporting modular parallelprogramming for distributed memory machines. It allows the user to specify dynamically an application topology, and the mpC programming environment uses this information in run time to provide the most efficient execution of the program on any particular distributed memory machine. The paper describes the features of mpC and its programming environment which allow to use them for developing libraries of parallel programs.
暂无评论