Serial to Parallel Matrix Inverse Algorithm using Gauss – Jordan Method
Fabery, Ramón O.
MetadataShow full item record
Abstract — In the past decade computers have become faster and more efficient, utilizing the technology available to put more transistors inside the same space; but when they can’t shrink the transistor any more is time to put more processors together to reduce the time of execution and have more than one processor making different instructions at the same time. This increment in throughput is possible using Message Passing Interface (MPI) a portable message-passing communication protocol that allows breaking apart the code and thus perform many instructions assigning to each processor the specific instruction to be executed making the process much faster. We are using a serial C programming language code program that calculates the inverse of a matrix using the Gauss-Jordan Method; then, we parallelize the same program using MPI. Key Terms — Matrix Inverse, Message Passing Interface, Multiples Cores, Parallel Computing.