2) Get the supplied code hw02_slow_code.cpp Identify possible optimizations. Implement optimizations and evaluate the results Assume vds and vi potentially contain values in the full dynamic range of their respective data types. 3) Implement a matrix-vector multiply from slide 10 of the DataAccessQntimization slide set. Benchmark performance for N 10, 100 and 1000 . Implement and benchmark both the outer loop unrolling and spatial blocking optimizations found on slides 11 and 12 2) Get the supplied code hw02_slow_code.cpp Identify possible optimizations. Implement optimizations and evaluate the results Assume vds and vi potentially contain values in the full dynamic range of their respective data types. 3) Implement a matrix-vector multiply from slide 10 of the DataAccessQntimization slide set. Benchmark performance for N 10, 100 and 1000 . Implement and benchmark both the outer loop unrolling and spatial blocking optimizations found on slides 11 and 12