Basic Linear Algebra Subprograms

수학노트
둘러보기로 가기 검색하러 가기

노트

  • The cuBLAS Library provides a GPU-accelerated implementation of the basic linear algebra subroutines (BLAS).[1]
  • The BLAS are a small core library of linear algebra utilities, which can be highly optimized for various architectures.[2]
  • The BLAS are used in a wide range of software, including LINPACK, LAPACK, and many other algorithms commonly in use today.[3]
  • BLAS 2 and BLAS 3 modules in SCSL are optimized and parallelized to take advantage of SGI's hardware architecture.[3]
  • SCSL also supports the C interface to the legacy BLAS set forth by the BLAS Technical Forum.[3]
  • The BLAS (Basic Linear Algebra Subprograms) are high quality "building block" routines for performing basic vector and matrix operations.[4]
  • Otherwise, an automatic optimized BLAS can be build, using the ATLAS package.[4]
  • This paper presents the implementation considerations and performance of the local BLAS, or BLAS local to each node of the system.[5]
  • The implications of implementing BLAS on distributed memory computers are considered in this light.[5]
  • Basic Linear Algebra Subroutines (BLAS) are routines that provide standard functions for basic vector and matrix operations.[6]
  • Refer to BLAS (Basic Linear Algebra Subprograms) for more information on the BLAS functions.[6]
  • In this paper, we only consider the single precision real and complex BLAS for vectors with positive strides.[7]
  • The PB‐BLAS consist of calls to the sequential BLAS for local computations, and calls to the BLACS for communication.[8]
  • Some of the linear algebra subprograms were designed in accordance with the Level 1 and Level 2 BLAS de facto standard.[9]
  • The vector-scalar linear algebra subprograms include a subset of the standard set of Level 1 BLAS.[9]
  • These subprograms include a subset of the standard set of Level 2 BLAS.[9]
  • Some of the matrix operation subroutines were designed in accordance with the Level 3 BLAS de facto standard.[9]
  • In this paper, we implement and evaluate the performance of some important BLAS operations on a matrix coprocessor.[10]
  • Fortunately, many applications are based on intensive use of Level-3 BLAS with small percentage of Level-1 and Level-2 BLAS.[10]
  • This paper describes a standard API for a set of Batched Basic Linear Algebra Subprograms (Batched BLAS or BBLAS).[11]
  • This design makes it easy to add further functionality to the sparse BLAS in the future.[12]
  • The full BLAS functionality for band-format and packed-format matrices is available through the low-level CBLAS interface.[13]
  • This interface corresponds to the BLAS Technical Forum’s standard for the C interface to legacy BLAS implementations.[13]
  • The library provides an interface to the BLAS operations which apply to these objects.[13]
  • LINPACK could use a generic version of BLAS.[14]
  • To gain performance, different machines might use tailored versions of BLAS.[14]
  • BLAS for a vector machine could use the machine's fast vector operations.[14]
  • Consequently, BLAS was augmented from 1984 to 1986 with level-2 kernel operations that concerned vector-matrix operations.[14]
  • Their beauty has always been that computer manufacturers have been encouraged to implement the BLAS as efficiently as possible.[15]
  • The vector BLAS are now called level-1 BLAS.[15]
  • This paper updates the ongoing BLAS effort, and summarizes what types of operations are now available.[15]
  • We now have flavors of BLAS for dense, banded, and sparse vector and matrix operations.[15]
  • This paper proposes adding a set of Level 3 BLAS, which would be used to perform matrix-matrix operations.[16]
  • Using the BLAS provides portability and ease of maintenance.[16]
  • The authors discuss the reasoning used in selecting the operations to be included in the Level 3 BLAS.[16]
  • An example illustrates how the Level 3 BLAS can be used to implement the Cholesky factorization as a block algorithm.[16]
  • (1979) published the original set of 38 BLAS.[17]
  • The IMSL BLAS collection includes these 38 subprograms plus additional ones that extend their functionality.[17]
  • (1988 and 1990) published extensions to this set, it is customary to refer to the original 38 as Level 1 BLAS.[17]
  • These are called the Level 2 BLAS (see).[17]

소스

  1. ↑ cuBLAS
  2. ↑ Basic Linear Algebra Subprograms
  3. ↑ 3.0 3.1 3.2 Chapter 2. Basic Linear Algebra Subprogram (BLAS) Routines
  4. ↑ 4.0 4.1 BLAS: Basic Linear Algebra Subprograms
  5. ↑ 5.0 5.1 Local Basic Linear Algebra Subroutines (LBLAS) for the CM-5/5E
  6. ↑ 6.0 6.1 Basic Linear Algebra Subroutines
  7. ↑ Basic linear algebra subprograms (BLAS) on the CDC CYBER 205
  8. ↑ PB‐BLAS: a set of parallel block basic linear algebra subprograms
  9. ↑ 9.0 9.1 9.2 9.3 Guide and Reference
  10. ↑ 10.0 10.1 Performance Evaluation of Basic Linear Algebra Subroutines on a Matrix Co-processor
  11. ↑ A Set of Batched Basic Linear Algebra Subprograms and LAPACK Routines
  12. ↑ An Overview of the Sparse Basic Linear Algebra Subprograms: The New Standard from the BLAS Technical Forum
  13. ↑ 13.0 13.1 13.2 BLAS Support — GSL 2.6 documentation
  14. ↑ 14.0 14.1 14.2 14.3 Basic Linear Algebra Subprograms
  15. ↑ 15.0 15.1 15.2 15.3 An updated set of basic linear algebra subprograms (BLAS)
  16. ↑ 16.0 16.1 16.2 16.3 A set of level 3 basic linear algebra subprograms
  17. ↑ 17.0 17.1 17.2 17.3 Basic Linear Algebra Subprograms

메타데이터

위키데이터

Spacy 패턴 목록

  • [{'LOWER': 'basic'}, {'LOWER': 'linear'}, {'LOWER': 'algebra'}, {'LEMMA': 'Subprograms'}]
  • [{'LEMMA': 'BLAS'}]
원본 주소 "https://wiki.mathnt.net/index.php?title=Basic_Linear_Algebra_Subprograms&oldid=51550"