Bo S

Bo S is a free and open-source software library for numerical linear algebra written in C++. It provides a rich collection of linear algebra routines optimized for high performance on multi-core and multi-processor systems.

Features

Bo S offers a wide range of features, including:

• Dense and sparse matrix operations
• Basic arithmetic operations (addition, subtraction, multiplication, etc.)
• Decomposition of matrices (LU, QR, Cholesky, etc.)
• Solvers for linear systems of equations
• Eigenvalue and eigenvector computations
• Random number generators for matrices and vectors
• Support for complex numbers

Performance

Bo S is designed to take advantage of modern hardware and software developments. It employs advanced algorithms and optimizations such as:

• Cache blocking
• Loop unrolling
• Vectorization
• Parallel processing with OpenMP
• Thread-level parallelism with TBB

As a result, Bo S can achieve high levels of performance on both desktop and supercomputer clusters.

Interface

Bo S follows a modern C++ interface with support for:

• Automatic memory management with smart pointers
• Expression templates for lazy evaluation
• User-defined types and operators

The library is compatible with a wide range of compilers and operating systems, including GCC, Clang, Visual Studio, macOS, and Linux.

Usage

Bo S is distributed under the permissive BSD license and can be freely used, modified, and redistributed. The library is available on GitHub and can be installed as a package or built from source.

Example code:

#include "bos/matrix.hpp"
#include "bos/decomposition.hpp"
#include "bos/solver.hpp"

int main() {
  bos::matrix<double> A = {{4, 12, -16},
                           {12, 37, -43},
                           {-16, -43, 98}};
  bos::vector<double> b = {1, 2, 3};

  // Solve Ax = b for x
  bos::lu_decomposition<double> lu(A);
  bos::vector<double> x = bos::lu_solver(lu, b);

  return 0;
}

Conclusion

Bo S is a powerful and versatile numerical linear algebra library for C++. With its high performance and modern interface, it is an excellent tool for scientists, engineers, and programmers who need to manipulate matrices and vectors in their applications.