A Novel Approach for Branch Prediction using SVM

Branch prediction accuracy is a crucial parameter in determining the amount of parallelism that can be exploited. Current advanced branch prediction techniques are able to attain correct predictions in most cases. Conventionally, the profiling information is used to predict the behavior of a branch. However, they are unsuccessful in predicting specific conditions like non linear branching and looping. This prevents them from achieving full accuracy. SVM have an inherent capability to analyze inputs to form meaningful relationships among them.