A5046544884- Matrix Theory and Algorithms
- Parallel Computing and Optimization Techniques
- Electromagnetic Scattering and Analysis
- Distributed and Parallel Computing Systems
- Advanced Numerical Methods in Computational Mathematics
- Advanced Optimization Algorithms Research
- Numerical methods for differential equations
- Advanced Data Storage Technologies
- Interconnection Networks and Systems
- Numerical Methods and Algorithms
- Service-Oriented Architecture and Web Services
- Model Reduction and Neural Networks
- Electromagnetic Simulation and Numerical Methods
- Embedded Systems Design Techniques
- Scientific Research and Discoveries
- Distributed systems and fault tolerance
- Cloud Computing and Resource Management
- Tensor decomposition and applications
- Context-Aware Activity Recognition Systems
- Human Mobility and Location-Based Analysis
- Mobile Ad Hoc Networks
- Statistical and numerical algorithms
- Neural Networks and Applications
- Computational Fluid Dynamics and Aerodynamics
- VLSI and FPGA Design Techniques
Prince Sultan University
2013-2023
Misr International University
2023
Purdue University West Lafayette
2011-2021
Helwan University
2016-2019
Center for Special Minimally Invasive and Robotic Surgery
2018
Liuyang City Maternal and Child Health Hospital
2018
Robotic Technology (United States)
2018
Robotic Research (United States)
2018
Azur Space Solar Power
2018
Wuhan University
2018
This report presents a methodology for measuring the performance of supercomputers. It includes 13 Fortran programs that total over 50,000 lines source code. They represent applications in several areas engi neering and scientific computing, many cases codes are currently being used by computational re search development groups. We also present PERFECT standard, set guidelines allow portability to types machines. Furthermore, we some measures method ology recording sharing results among...
article Free AccessOn Stable Parallel Linear System Solvers Authors: A. H. Sameh Department of Computer Science, University Illinois, Urbana, IL ILView Profile , D. J. Kuck Authors Info & Claims Journal the ACMVolume 25Issue 1Jan. 1978 pp 81–91https://doi.org/10.1145/322047.322054Published:01 January 1978Publication History 189citation1,045DownloadsMetricsTotal Citations189Total Downloads1,045Last 12 Months50Last 6 weeks7 Get Citation AlertsNew Alert added!This alert has been successfully...
More and more scientists engineers are becoming interested in using supercomputers. Earlier barriers to these machines disappearing as software for their use improves. Meanwhile, new parallel supercomputer architectures emerging that may provide rapid growth performance. These systems a large number of processors with an intricate memory system is both hierarchical; they will require even advanced software. Compilers restructure user programs exploit the machine organization seem be...
The reasons for the creation of Illiac IV are described and history project is recounted. architecture or hard-ware structure discussed--the array an processor with a specialized control unit (CU) that can be viewed as small stand-alone computer. software strategy in terms current user habits needs. Brief descriptions given systems itself, its history, major lessons learned during development. Some ideas future development suggested. Applications discussed evaluating function f(x)...
Scientific and engineering research is becoming increasingly dependent upon the development implementation of efficient parallel algorithms on modern high-performance computers. Numerical linear algebra an indispensable tool in such this paper attempts to collect describe a selection some its more important algorithms. The purpose review current status provide overall perspective for solving dense, banded, or block-structured problems arising major areas direct solution systems, least...
An algorithm for computing a few of the smallest (or largest) eigenvalues and associated eigenvectors large sparse generalized eigenvalue problem $Ax = \lambda Bx$ is presented. The matrices A B are assumed to be symmetric, haphazardly sparse, with being positive definite. treated as one constrained optimization an inverse iteration developed which requires solution linear algebraic systems only accuracy demanded by given subspace. rate convergence method established, technique improving it...
Many existing algorithms for obtaining the eigenvalues and eigenvectors of matrices would make poor use such a powerful parallel computer as ILLIAC IV. In this paper, Jacobi’s algorithm real symmetric or complex Hermitian matrices, Jacobi-like nonsymmetric developed by P. J. Eberlein, are modified so to achieve maximum efficiency computations.
article Free Access Share on CEDAR: a large scale multiprocessor Authors: Daniel Gajski University of Illinois at Urbana-Champaign Urbana-ChampaignView Profile , David Kuck Duncan Lawrie Ahmed Sameh Authors Info & Claims ACM SIGARCH Computer Architecture NewsVolume 11Issue 1March 1983 pp 7–11https://doi.org/10.1145/859526.859527Online:01 March 1983Publication History 33citation221DownloadsMetricsTotal Citations33Total Downloads221Last 12 Months10Last 6 weeks1 Get Citation AlertsNew Alert...
Linear algebra algorithms based on the BLAS or ex tended do not achieve high performance mul tivector processors with a hierarchical memory system because of lack data locality. For such machines, block linear must be implemented in terms matrix-matrix primitives (BLAS3). Designing ef ficient for these architectures requires analysis behavior and resulting as func tion certain parameters. The identify limits improvement possible via blocking any contradictory trends that require trade-off...
article Free Access Share on Practical Parallel Band Triangular System Solvers Authors: S. C. Chen Department of Computer Science, University Illinois at Urbana-Champaign, Urbana, IL ILView Profile , D. J. Kuck A. H. Sameh Authors Info & Claims ACM Transactions Mathematical SoftwareVolume 4Issue 3Sept. 1978 pp 270–277https://doi.org/10.1145/355791.355797Published:01 September 1978Publication History 80citation447DownloadsMetricsTotal Citations80Total Downloads447Last 12 Months20Last 6 weeks3...
Three conjugate gradient accelerated row projection (RP) methods for nonsymmetric linear systems are presented and their properties described. One method is based on Kaczmarz's has an iteration matrix that the product of orthogonal projectors; another Cimmino's sum projectors. A new RP method, which requires fewer matrix-vector operations, explicitly reduces problem size, error reducing in two-norm, consistently produces better solutions than other algorithms, also introduced. Using...
A multiprocessor algorithm for finding few or all eigenvalues and the corresponding eigenvectors of a symmetric tridiagonal matrix is presented. It pipelined variation EISPACK routines—BISECT TINVIT which consists three steps: isolation, extraction-inverse iteration, partial orthogonalization. Multisections are performed isolating in given interval, while bisection Zeroin method used to extract these isolated eigenvalues. After have been computed by inverse modified Gram-Schmidt...
In this paper we present alternative formulations of the algorithms Chen and Kuck [IEEE Trans, Computers (1975)]. We also give a detailed error analysis, showing that if $\tilde x$ is computed solution triangular system $Lx = f$, then it satisfies equation $(L + \delta L)\tilde x f$ where $\| {\delta L} \| \leqq O(n^2 \log n)\varepsilon \kappa ^2 (L)\| L \|$. Here $\kappa (L)$ condition number L, \cdot \|$ denotes $\infty $-norm, $\varepsilon $ unit roundoff.
article Free Access Share on The computation and communication complexity of a parallel banded system solver Authors: D H. Lawrie University Illinois at Urbana-Champaign Urbana-ChampaignView Profile , A Sameh Authors Info & Claims ACM Transactions Mathematical SoftwareVolume 10Issue 2June 1984 pp 185–195https://doi.org/10.1145/399.401Published:18 May 1984Publication History 60citation490DownloadsMetricsTotal Citations60Total Downloads490Last 12 Months28Last 6 weeks6 Get Citation AlertsNew...
We show that if the size of tridiagonal matrix in any given iteration is n, then parallel QR algorithm requires 0(log <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> n) steps with 0(n) processors per and no square roots. This results a speedup 0(n/log over sequential an efficiency 0(1/log n). also give error analysis triangular system solvers used each iteration.