A set of level 3 basic linear algebra subprograms

By Jack J. Dongarra, J. D. Croz, S. Hammarling, Iain S. Duff. 1990

Subfields: Numerical Methods & Scientific Computing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Numerical Methods & Scientific Computing

Citations in the corpus

Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation (2020, unknown): “Furthermore, if per-tensor quantization is used for both arguments, a single ﬂoating-point multiplier is needed and is part of the GEMM API (often referred to as alpha) in BLAS libraries [11].”
An overview of the Trilinos project (2005, methodology): “By using partial specialization of templates, we are basing Tpetra on established libraries such as the BLAS [Lawson et al. 1979; Dongarra et al. 1988; Dongarra et al. 1990] and LAPACK [Anderson et al.”
Automated empirical optimizations of software and the ATLAS project (2001, background): “One of the main performance kernels of linear algebra has traditionally been a standard API known as the BLAS (Basic Linear Algebra Subprograms) [14, 17, 6, 7, 5] This API is supported by hand-tuned efforts of many hardware vendors, and thus provides a good first target for ATLAS, as there is both a”
Sourcebook of parallel computing (2003, methodology): “Since their inception, the use of de facto standards like the BLAS [283, 284] has been a means of achieving portability and efficiency for a wide range of kernel scientific computations.”
A data locality optimizing algorithm (1991, methodology): “Blocking [9] or tiling [18] is a well-known technique that improves the data locality of numerical algorithms [1, 6 , 7, 12, 13].”
A two-stage ensemble Kalman filter for smooth data assimilation (2008, methodology): “The dominant operations are full matrix-matrix operations efficiently implemented in the Level 3 BLAS (Dongarra et al. 1990) and LAPACK (Anderson et al. 1999) routines, and also readily parallelizable by SCALAPACK (Blackford et al. 1997).”
An update to the Horizontal Wind Model (HWM): The quiet time thermosphere (2015, methodology): “EMPIRICAL MODEL OF THERMOSPHERIC WINDS 308 This is achieved with the sparse Basic Linear Algebra Subprogram matrix library functions [Dongarra et al., 1990] and the iterative linear least squares optimal estimation procedure described by Rodgers [2000].”
Introduction to the HPCChallenge Benchmark Suite (2004, methodology): “The HPC Challenge benchmark consists at this time of 7 performance tests: HPL [3], STREAM [4], RandomAccess , PTRANS , FFT (implemented using FFTE [5]), DGEMM [6, 7] and b eff (MPI latency/bandwidth test) [8, 9, 10].”
Minimizing development and maintenance costs in supporting persistently optimized BLAS (2005, methodology): “To handle this problem, a standard API of performance-critical linear algebra kernels was created, called the Basic Linear Algebra Subprograms (BLAS) [1–5], which provides such linear algebra kernels as matrix multiply, triangular solve, etc.”
Computing the Action of the Matrix Exponential, with an Application to Exponential Integrators (2011, background): “The advantage of working with B is the ability to invoke level 3 BLAS [6] [7], which should lead to faster execution.”
Interpolating implicit surfaces from scattered surface data using compactly supported radial basis functions (2001, methodology): “We use a direct (LU) sparse matrix solver [ 4 ] to find the solution to”
A column pre-ordering strategy for the unsymmetric-pattern multifrontal method (2004, methodology): “This BLAS increases the performance of UMFPACK4 by about 50% for large sparse matrices on this computer, as compared with the ATLAS 3.4.1 BLAS [Whaley et al. 2001].”
Linear algebra operators for GPU implementation of numerical algorithms (2003, methodology): “…operators as available in widespread linear algebra libraries, i.e. the BLAS (Basic Linear Algebra Subprogram) library [Dongarra et al. 1988; Dongarra et al. 1990], by GPU implementations, thus enabling more general linear algebra packages to be implemented on top of these implementations,…”
TMB: Automatic Differentiation and Laplace Approximation (2015, methodology): “Its design takes advantage of the following high performing and well maintained software tools: R , CppAD , Eigen , BLAS , and CHOLMOD .”