A fundamental and ubiquitous difficulty of systems biology is identifying relevant model parameters. A genetic network model of the biological clock of Neurospora crassa that is quantitatively consistent with the available RNA and protein profiling data was proposed. However, the oscillating nature of biological models poses more challenge for identifying model parameters due to the high dimensional complex search space and computational cost of numerically solving ODEs. In this work, an Evolutionary Algorithm leveraging the GPU architecture is proposed. Our implementation identified promising model parameters with a speedup of two orders of magnitude compared to the CPU implementation.