China
Africa
Explore chapters and articles related to this topic
Force-System Resultants and Equilibrium
Published in Richard C. Dorf, The Engineering Handbook, 2018
LINPACK is a collection of programs concerned with direct methods for general (or full) symmetric, symmetric positive definite, triangular, and tridiagonal matrices. There are also programs for least squares problems, along with the QR algorithm for eigensystems and the singular value decompositions of rectangular matrices. The programs are intended to be completely machine independent, fully portable, and run with good efficiency in most computing environments. The LINPACK User’s Guide by Dongarra et al. is the basic reference.
Using Performance Metrics to Select Microprocessor Cores for IC Designs
Published in Louis Scheffer, Luciano Lavagno, Grant Martin, EDA for IC System Design, Verification, and Testing, 2018
LINPACK is a collection of FORTRAN subroutines that analyze and solve linear equations and linear least-squares problems. Jack Dongarra assembled the LINPACK collection of linear algebra routines at the Argonne National Laboratory, Argonne, IL. The first versions of LINPACK existed in 1976 but the first users’ guide was published in 1977 [5,6]. The package solves linear systems whose matrices are general, banded, symmetric-indefinite, symmetric positive definite, triangular, and tridiagonal square. In addition, the package computes the QR and singular value decompositions of rectangular matrices and applies them to least-squares problems. The LINPACK routines are not strictly speaking a benchmark, but they exercise a computer’s floating-point capabilities and, as of 2005, Dongarra maintains an online list of computer systems and their LINPACK performance results at www.netlib.org.
Linear Algebra Problems
Published in Dingyü Xue, YangQuan Chen, Scientific Computing with MATLAB®, 2018
Although many readers might not be able to quickly compute the determinant of a given 3 × 3 or even 4 × 4 matrix by hand, one should not feel bad about this inability. We leave this low-level computation to computer mathematics languages, such as MATLAB. In fact, many computer mathematics languages, such as MATLAB, originated from the early research of numerical linear algebra. For instance, the well-known EISPACK package [1] focused on the computation of eigen-systems of matrices. Another well-known package, LINPACK [2], was developed to solve general linear algebra problems using numerical algorithms. With the development of computer science, matrix computations are now no longer restricted to numerical computations. Analytical solutions can also be found for many linear algebra problems. Successful computer mathematics languages such as Mathematica, Maple and the Symbolic Math Toolbox of MATLAB can be used to analytically solve certain problems in linear algebra.
A Container-Based Technique to Improve Virtual Machine Migration in Cloud Computing
Published in IETE Journal of Research, 2022
Aditya Bhardwaj, C. Rama Krishna
In the second subcategory, the tools explored are: Linpack and UnixBench. Linpack is a benchmark software which seeks to measure the performance of the running system by generating linear equations and solves using lower–upper decomposition. We tested this workload with a random matrix M of size N=1000. Another benchmarking tool utilized is UnixBench, which is used to evaluate the system performance in single and multi-threading environment. Finally, we consider Sysbench as a CPU prime number compute-intensive benchmark, with a default parameter of CPU-max-prime checked = 20,000.