Branch Prediction in CPU Pipelining

Branch prediction is a prominent technology used in pipelined CPUs to help reduce the delay experienced in preventing control hazards. There are many different branch prediction techniques and algorithms proposed ranging from static to dynamic, each of which varies in terms of accuracy in predicting incoming branch directions. In this research paper, we use a simulation to test five different branch prediction algorithms and analyze the values obtained from their trace files, as well as the clock cycle times of each run. From the results, the static algorithms performed the worst and the dynamic algorithm based on a neural network performed the best. However, the ideal branch predictor cannot be determined solely by these results.