ECEC-412/621
Project 2: Implementing a perceptron-based branch predictor and evaluating
its performance using ML workloads.
1 Perceptron-based Branch Predictor
1. Read through Supplement One and implement a perceptron-based branch predictor.
2. Evaluate your branch predictor with the AI workloads and find out the best performing configurations.
3. Compare your branch predictor against the two-bit local, tournament and gShare predictor. Does the
perceptron-based branch predictor out-perform them?
2 Submission
1. Report how you design the perceptron-based branch predictor with a diagram and pseudo-code.
2. Summarize the experiment in Section 1.
3. Compile all the above information in a single PDF file.
4. Cleaned source codes.
5. Zip above all and submit through Bblearn.
1
