Abstract: The performance evaluatation of ground layer adaptive opitcs (GLAO) is helpful for system design and optimization. The turbulence distribution and the layout of the guide stars (GSs) are the main factors affecting GLAO performance. Considering the impact of the turbulence distribution and the layout of GSs, the performance of GLAO was analysed and evaluated by comparing spatial frequency spectrum filtering theory and Monte Carlo simulation. The results show that the conclusions of the two methods are clearly consistent with an error margin of less than 4.6%. Spatial frequency spectrum filtering simplifies the system into a linear model and it is simpler and faster for the calculation, which is convenient for discovering the characteristic rules. However, the accuracy is slightly low if considering the noise and error in a real system. In addition, this method is suitable for analysing the system performance with a symmetrical GS layout. The Monte Carlo method is better for simulating the system running state in detail with a random GS layout. The brief results of the system performance analysis are given in the end by combining the two methods. The study will be useful for the system design and optimization of future GLAOs.