BLAS
E440649
BLAS (Basic Linear Algebra Subprograms) is a standardized collection of low-level routines for performing common linear algebra operations such as vector and matrix multiplication, widely used as a performance-optimized foundation in scientific computing.
All labels observed (1)
| Label | Occurrences |
|---|---|
| BLAS canonical | 5 |
Statements (47)
| Predicate | Object |
|---|---|
| instanceOf |
application programming interface
ⓘ
numerical linear algebra library specification ⓘ software standard ⓘ |
| abbreviation | BLAS NERFINISHED ⓘ |
| defines | standard interfaces for linear algebra kernels ⓘ |
| designedFor |
high performance on a wide range of architectures
ⓘ
portability across hardware platforms ⓘ |
| field |
high‑performance computing
ⓘ
numerical linear algebra ⓘ scientific computing ⓘ |
| fullName | Basic Linear Algebra Subprograms NERFINISHED ⓘ |
| hasImplementation |
ATLAS
NERFINISHED
ⓘ
Apple Accelerate framework NERFINISHED ⓘ IBM ESSL NERFINISHED ⓘ Intel Math Kernel Library NERFINISHED ⓘ Netlib reference BLAS NERFINISHED ⓘ OpenBLAS NERFINISHED ⓘ cuBLAS NERFINISHED ⓘ |
| hasLevel |
Level 1 BLAS
NERFINISHED
ⓘ
Level 2 BLAS NERFINISHED ⓘ Level 3 BLAS NERFINISHED ⓘ |
| languageBinding |
C
NERFINISHED
ⓘ
C++ ⓘ Fortran NERFINISHED ⓘ Python (via wrappers such as NumPy and SciPy) ⓘ |
| layer | low‑level numerical software ⓘ |
| operationType |
matrix‑matrix operations
ⓘ
matrix‑vector operations ⓘ vector operations ⓘ vector‑vector operations ⓘ |
| optimizationTarget |
cache efficiency
ⓘ
parallel execution ⓘ vectorization ⓘ |
| property | interface is standardized while implementations may be vendor‑optimized ⓘ |
| purpose |
provide standardized low‑level routines for linear algebra operations
ⓘ
serve as a performance‑optimized foundation for higher‑level numerical libraries ⓘ |
| standardizedBy | Netlib community NERFINISHED ⓘ |
| supportsDataType |
double‑precision complex
ⓘ
double‑precision real ⓘ single‑precision complex ⓘ single‑precision real ⓘ |
| typicalUseCase |
dense linear algebra computations
ⓘ
machine learning workloads ⓘ numerical simulations ⓘ |
| usedAs |
backend for linear algebra in many programming environments
ⓘ
building block for LAPACK ⓘ building block for many scientific computing libraries ⓘ |
Referenced by (5)
Full triples — surface form annotated when it differs from this entity's canonical label.