The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Some memory writes have the particular behaviour of not modifying memory since the value they write is equal to the value before the write. These kind of stores are what we call Redundant Stores. In this paper we study the behaviour of these particular stores and show that a significant saving on memory traffic between the first and second level caches can be avoided by exploiting this feature. We...
The paper is devoted to the problems of presenting results obtained during the numerical simulation. Two different approaches of data visualization: through 2D and 3D objects are described. 3D-visualization requires more computational resources than 2D, especially when the data volumes are large. So the best way for quick getting necessary data during the numerical simulation is to use supercomputers...
In the paper, a method is proposed for optimizing parallelization of loops whose dependence graphs are acyclic. The idea of loop distribution as a method for data dependence synchronization is presented. Next, the loop distribution optimization problem is posed and solved with the technique of optimal dependence folding. Theoretical considerations are accompanied by experimental results from applying...
Maximum clique and independent set problems are classical NP-full optimization problems, the solutions of which are difficult to obtain from conventional methods. Hopfield network in neural network, which simulates the partial functions of a human brain through the ultra-large scale parallel computation, has been proven to have potentials in solving these problems in a reasonable period of time. The...
As the scientific data become larger and larger, a method is needed to store and retrieve it in time- and cost-efficient way. NCSA HDF is a data/programming model widely used for storing and sharing scientific data. We present an approach to store large volumes of structured scientific data in HDF format on tertiary storage.
A model of the http traffic generated by a community of users connected to the Internet via a proxy cache is described. The model reproduces Internet traffic realistically and is used as input to the Internet cache simulation models developed by British Telecom research laboratories.
This paper presents the motivation and insight into modifications and extensions of two performance monitoring tools—PATOP and TATOO—to make them work with message passing applications based on PVM and MPI. Also discussed are the concepts of porting the tools to the OCM monitoring environment, the structure of the modified tools and the extensions made.
The OCM is a universally usable, distributed on-line monitoring system currently implemented for the PVM programming library. Due to the growing use of MPI in parallel programming, there is a need to enhance the OCM in order to support MPI applications development. This paper presents approaches to solve the problems of the mpichoriented start-up mechanism and profiling MPI library calls in the OCM...
In this paper we present a functionality and a structure of a symbol table manager for debugging HPF programs. The manager retrieves information from executable files generated by different Fortran 77/90 and HPF compilers and converts it into an internal representation independent of file format, which is accessed by the debugging system.
This paper presents a data flow analysis framework which plays a key role in identifying opportunities for communication optimization in tuplespace parallel programs.
A new method, based on the functional formulation of the theory of quantum mechanism multichannel scattering for three-body collinear systems is proposed. Based on intrinsic properties of scattering system the numerical task was divided into independent subtasks and the parallel algorithm for numerical computations was developed and tested on massive-parallel systems Parsystec CC/16 and SPP-1600....
In this paper, we discuss the parallel sparse Cholesky factorization problem for distributed memory multiprocessor systems. Although there are already several articles for this problem, most of them only focus on the processor allocation. We discuss the processor allocation and the task scheduling issues simultaneously to minimize the overall parallel execution time. Several existing methods are compared...
Message Passing Interface[2] is the de facto standard for multi-computer and cluster message passing. In this paper we explore a new paradigm of high performance distributed computing by implementing a message passing interface over HTTP[1]. This provides a platform independent implementation of MPI and also develops a base for web based computation to achieve global parallel processing. The conventional...
A distributed vision network is proposed to tackle industrial packaging inspection. The system consists of independent networked inspection stations able to address efficiently parallel inspection tasks such as product identification, character verification, tag inspection and content & packaging quality control at a high production speed. Existing and innovative inspection algorithms such as...
Taking as a starting point for FPGAs design and efficient bit-level systolic algorithm facilitates the design process but does not automatically guarantee the most efficient hardware solution. We demonstrate on an example of Montgomery exponentiation a role of partitioning in mapping of linear systolic arrays onto Xilinx XC6000 FPGAs.
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.