Skip to content

Commit d1bc98e

Browse files
committed
Update README
1 parent 666a9d7 commit d1bc98e

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

README.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
# Fastor
2-
**Fastor** [**FA**st **S**IMD op**T**imised tens**OR** algebra framework]: is a smart high performance stack-based tensor (multi-dimensional array) library written in modern C++ [C++11/14/17] with powerful in-built tensor algebraic functionalities (tensor contraction, permutation, reductions, orthogonal and special tensor groups and many more). Designed as a generic multi-dimensional tensor algebra library, Fastor also incorporates domain specific features for tensor contraction algorithms typically arising in classical mechanics, in particular, in finite element analysis of nonlinear solids, fluids and coupled continua. There are multiple paradigms that Fastor exploits:
2+
**Fastor** [**FA**st **S**IMD op**T**imised tens**OR** algebra framework]: is a smart stack-based high performance tensor (multi-dimensional array) library written in modern C++ [C++11/14/17] with powerful in-built tensor algebraic functionalities (tensor contraction, permutation, reductions, special tensor groups and much more). Designed as a generic multi-dimensional tensor algebra library, Fastor also incorporates domain specific features for tensor contraction algorithms typically arising in classical mechanics, in particular, in finite element analysis of nonlinear solids, fluids and coupled continua. There are multiple paradigms that Fastor exploits:
33

44
- **Operation minimisation/Low FLOP/Complexity reducing Algorithms:** Fastor relies on an extremely smart and domain-aware Expression Template (ET) engine that can not only perform lazy evaluations and operator chaining but can also perform sophisticated mathematical transformation or *compile time* graph optimisation or both to reduce the complexity of evaluation of expressions by orders of magnitude. Some of these functionalities are almost non-existing in other available C++ ET linear algebra frameworks.
55
- **SIMD/Data parallelism/Stream computing** Fastor utilises explicit SIMD (SSE/SSE2/SSE3/SSE4/AVX/AVX2/AVX512/FMA) instructions

0 commit comments

Comments
 (0)