DSP processor growth is phenomenal and continues to grow rapidly, but general-purpose microprocessors have entered the multimedia and signal processing oriented stream by adding DSP functionality to the instruction se...
详细信息
We study the system performance affected by Degree of Declustering (DD) and the additional overheads in shared-nothing parallel transaction processing systems withthe widely used two-phased locking (2PL) concurrency ...
详细信息
ISBN:
(纸本)3540663630
We study the system performance affected by Degree of Declustering (DD) and the additional overheads in shared-nothing parallel transaction processing systems withthe widely used two-phased locking (2PL) concurrency control method by simulation. We show that for the environment of high resource contention, a large DD is beneficial. For a system in the environment of low resource contention, if we have infinite frontend resource, a large DD is beneficial;otherwise a large DD would degrade system performance. It is shown that as the level of the resource contention in disks decreases, the additional overhead of parallelism increases. Furthermore, it is shown that the functions attached to frontend nodes affect system performance to a significant degree for shared-nothing parallel TP systems. these results provide insights for improving performance of shared-nothing parallel transaction processing systems.
Message-passing is often used to implement parallel programs to run on workstation clusters. However, writing message-passing programs is a difficult and error prone task. In this paper we describe a graphical interfa...
详细信息
the SCOOPP (Scalable Object Oriented parallel Programming) system is an hybrid compile and run-time system, that extracts parallelism, supports explicit parallelism and dynamically serialises parallel tasks in excess,...
详细信息
We have studied a preconditioning technique for Krylov subspace methods on a fluid dynamics problem in 2-D. By discretizing;the time-dependent Euler equations with a finite volume method in space and using the trapezo...
详细信息
ISBN:
(纸本)3540664432
We have studied a preconditioning technique for Krylov subspace methods on a fluid dynamics problem in 2-D. By discretizing;the time-dependent Euler equations with a finite volume method in space and using the trapezoidal rule in time, we get a nonlinear system which is solved using a Newton-Krylov method. We precondition the linear iterates using a parallel semi-Toeplitz preconditioner to reduce the number of iterations. the experiments show a substantial reduction in the number of iterations required for convergence.
We present a cost model that relies on an asynchronous and structured parallel execution model for message passing. We show that it is possible to define a complexity function for programs that yields a symbolic date ...
详细信息
ISBN:
(纸本)3540664432
We present a cost model that relies on an asynchronous and structured parallel execution model for message passing. We show that it is possible to define a complexity function for programs that yields a symbolic date for each communication event. By ordering these dates, an upper bound for the network load can be computed. In contrast to classical approaches this cost computation can handle asynchronism for message passing and communication/computation overlap.
Many advanced scientific applications are heterogeneous and multidisciplinary in nature, consisting of multiple, independent modules. Such applications require efficient means of coordination for their program units. ...
详细信息
ISBN:
(纸本)3540664432
Many advanced scientific applications are heterogeneous and multidisciplinary in nature, consisting of multiple, independent modules. Such applications require efficient means of coordination for their program units. the programming language Opus was designed recently to assist in coordinating the execution of multiple, independent program modules. In this paper we address the problem of how to compile an Opus program such that it can be efficiently executed on a broad class of machines.
Most methods for computing the singular value decomposition (SVD) first bidiagonalize the matrix. the ScaLAPACK implementation of the blocked reduction of a general dense matrix to bidiagonal form performs about one h...
详细信息
ISBN:
(纸本)3540664432
Most methods for computing the singular value decomposition (SVD) first bidiagonalize the matrix. the ScaLAPACK implementation of the blocked reduction of a general dense matrix to bidiagonal form performs about one half of the operations with BLAS3. If we subdivide the task into two stages dense --> banded and banded --> bidiagonal, we can increase the portion of matrix-matrix operations and expect higher performance. We give an overview of different techniques for the first stage. this note summarizes the results of [9, 10].
this paper presents the rationale and some implementation aspects of the VOTE communication support system. VOTE supports a global address space by means of a software-implemented virtual shared memory (VSM). VSM cons...
详细信息
Performance estimates of a parallel computer during sparse matrix factorisation aid in the identification of overheads and the tuning of software. this paper proposes a technique which allows the computer parameters o...
详细信息
暂无评论