Results 291 to 300 of about 21,055,470 (353)

Exploitation of control parallelism in data parallel algorithms

Proceedings Frontiers '95. The Fifth Symposium on the Frontiers of Massively Parallel Computation, 2002
This paper considers the matrix decomposition A=LDL/sup T/, as a vehicle to explore the improvement in performance obtainable through the execution of multiple streams of control on SIMD architectures. Several methods for partitioning the SIMD array are considered.
V. Garg, D.E. Schimmel
openaire   +2 more sources

PROBLEMS WITH DATA PARALLELISM

Parallel Processing Letters, 2001
The gradual evolution of language features and approaches used for the programming of distributed memory machines underwent substantial advances in the 1990s. One of the most promising and widely praised approaches was based on data parallelism and resulted in High Performance Fortran. This paper reports on an experiment using that approach based on a
Carol Phillips, Ronald H. Perrott
openaire   +2 more sources

Parallel simulation of data parallel programs

1996
Accurate simulations of parallel programs for large datasets can often be slow; parallel execution has been shown to offer significant potential in reducing the execution time of many discrete-event simulators. In this paper, we describe the design and implementation of a parallel simulator called DPSIM that simulates the execution of data parallel ...
Sundeep Prakash, Rajive Bagrodia
openaire   +2 more sources

Loop Parallelism Maximization for Multimedia Data Processing in Mobile Vehicular Clouds

IEEE Transactions on Cloud Computing, 2019
Mobile vehicular cloud has become popular with the rapid development of cloud computing and mobile computing. Nested loops are usually the most critical part in multimedia and high performance Digital Signal Processing (DSP) systems which are widely used
Meikang Qiu, Wenyun Dai, A. Vasilakos
semanticscholar   +1 more source

Parallel Image Processing with the Block Data Parallel Architecture

IBM Journal of Research and Development, 1996
Many digital signal and image processing algorithms can be speeded up by executing them in parallel on multiple processors. The speed of parallel execution is limited by the need for communication and synchronization between processors. In this paper, we present a paradigm for parallel processing that we call the block data flow paradigm (BDFP).
Douglas S. Reeves, Winser E. Alexander, Clay Gloster   +2 more
openaire   +3 more sources

Data parallel algorithms

Communications of the ACM, 1986
Parallel computers with tens of thousands of processors are typically programmed in a data parallel style, as opposed to the control parallel style used in multiprocessing. The success of data parallel algorithms—even on problems that at first glance seem inherently serial—suggests that this style of programming has much wider applicability than was ...
W. Daniel Hillis, Guy L. Steele
openaire   +2 more sources

PipeDream: generalized pipeline parallelism for DNN training

Symposium on Operating Systems Principles, 2019
DNN training is extremely time-consuming, necessitating efficient multi-accelerator parallelization. Current approaches to parallelizing training primarily use intra-batch parallelization, where a single iteration of training is split over the available ...
D. Narayanan, A. Harlap, Amar Phanishayee, Vivek Seshadri, Nikhil R. Devanur, G. Ganger, Phillip B. Gibbons, M. Zaharia   +7 more
semanticscholar   +1 more source

Data Parallelism, Control Parallelism, and Related Issues

2000
This Chapter focuses on the differences between control parallelism and data parallelism, which are important to understand the discussion about parallel data mining in later Chapters of this book. After an introduction to control and data parallelism, we discuss the effect of exploiting these two kinds of parallelism in three important issues, namely ...
Simon H. Lavington, Alex A. Freitas
openaire   +2 more sources

Data Parallel Fortran

[Proceedings 1992] The Fourth Symposium on the Frontiers of Massively Parallel Computation, 1992
The authors present Data Parallel Fortran (DPF), a set of extensions to Fortran aimed at programming scientific applications on a variety of parallel machines. DPF portrays a global name space to programmers and allows programs to be written in a clear, data-parallel style.
P.M. Elustondo, J.L.C. Sanz, G.A. Alvarez, O.J. Nestares, J.A.S. Avalos, C.-T. Ho, L.A. Vazquez   +6 more
openaire   +2 more sources

On Parallelizing the MRRR Algorithm for Data-Parallel Coprocessors

2010
The eigenvalues and eigenvectors of a symmetric matrix are of interest in a myriad of applications. One of the fastest and most accurate numerical techniques for the eigendecomposition is the Algorithm of Multiple Relatively Robust Representations (MRRR), the first stable algorithm that computes the eigenvalues and eigenvectors of a tridiagonal ...
Paolo Bientinesi, Christian Lessig
openaire   +2 more sources