Parallel implementation of the Ising Model using C++ with the support of the Intel MKL library for efficient random number generation and OpenMP for parallelization. The Ising Model is a mathematical model of ferromagnetism in statistical mechanics, and this implementation aims to leverage modern computing resources to perform large-scale simulations efficiently.
- Random Number Generation: Utilizes the Intel MKL library for random number generation.
- Parallel Processing: Employs OpenMP to parallelize random number generation and lattice updates.
- C++ Compiler: A compliant C++ compiler (e.g., g++, clang++).
- Intel MKL Library: Intel Math Kernel Library for random number generation.
- OpenMP: OpenMP support for parallel processing.