This project implements a high-performance matrix multiplication tool in Rust by leveraging concurrency and parallelism. It can calculate the product of two matrices with arbitrary dimensions.

Accepts two matrices of any size as input Spawns a thread pool to perform concurrent multiplication Divides each matrix into smaller blocks Each thread multiplies a block from matrix A with a block from matrix B Accumulates the partial results into the final output matrix Matrix blocking and threading improves CPU utilization

Built using Rust and its native threads and synchronization primitives Creates a fixed size thread pool on startup Blocks input matrices into smaller sub-matrices based on number of threads Each thread picks up a block-wise multiplication job from a queue Results are placed in the output matrix through synchronized access

Concurrent design exploits parallelism in multi-core CPUs Matrix blocking improves cache utilization compared to naive multiplication Achieves near linear speedup with increase in number of threads Significantly faster than single-threaded implementation, This tool shows the performance benefits of concurrency and matrix decomposition for parallel workloads like matrix math. The project can be enhanced by exploring GPU computation and SIMD optimizations.