REAL-WORLD optimization problems are oftenns2 projects in nigeria complex and NP-hard. Their modeling is continuously evolving in ns2 projects in nigeria terms of constraints and objectives, and their resolution ns2 projects in nigeria is CPU time consuming. Although near-optimal algorithms such as metaheuristics make it possible to ns2 projects in nigeria reduce the temporal complexity of their resolution, ns2 projects in nigeria they fail to tackle large problems satisfactorily. GPU computing has recently ns2 projects in nigeria been revealed effective to deal with timeintensive problems Our ns2 project in nigeria challenge is to rethink the design of metaheuristics on GPU for solving ns2 projects in nigeria large-scale complex problems with a view to high effectiveness ns2 projects in nigeria and efficiency. Metaheuristics are based on the iterative improvement of either single solution or a population ns2 projects in nigeria of solution of a given optimization problem. In this ns2 projects in nigeria paper, we focus on the first category, i.e., local search metaheuristics. This class ns2 projects in nigeria of algorithms handles a single solution which is iteratively improved ns2 projects in nigeria by exploring its neighborhood in the solution space. The neighborhood structure ns2 projects in nigeria depends on the solution encoding which could mainly be a binary encoding, a vector of discrete values, a permutation, or a vector of real values. For years, ns2 projects in nigeria the use of GPU accelerators was to graphics applications. Recently, ns2 projects in nigeria their use has been extended to other ns2 projects in nigeria application domains ns2 projects in nigeria thanks to the publication of the Compute Unified Device Architecture (CUDA) ns2 project in nigeria development toolkit that allows GPU programming in C-like ns2 project in nigeria language. In some areas such as numerical computing we are now ns2 project in nigeria witnessing the proliferation of software libraries such as CUBLAS for GPU. However, in other areas such as ns2 project in nigeria combinatorial optimization, the spread ns2 project in nigeria of GPU does not occur at the same pace. Indeed, there only exists few ns2 project in nigeria research works related to evolutionary algorithms on GPU ns2 project in nigeria Nevertheless, parallel combinatorial optimization.