Parallelizations of the Error Correcting Code Problem

In this work, an exact approach to solve the Error Correcting Code problem is presented. For the implementation, the Branch and Bound skeleton of the MaLLBa library has been applied. This tool provides a hierarchy of C++classes which must be adapted to the specific requirements of the problem. Then, it generates two parallel solvers: one based on the message passing paradigm and other designed on the basis of a shared memory model. For both parallel proposals the sequential algorithm follows the same principles. Note that with a single and simple specification of the problem, the tool gives the user two different parallel approaches. Computational results obtained with the openmpand mpitools are shown.