Consider Proten String Matching Code. It takes 4 days exeution time on current machine. 20% percent of the time doing integer instructions and 35% percent of time doing I/O. Which is the better tradeoff? 1) Compiler optimization that reduces number of integer instructions by 25% (assume each integer instruction takes the same amount of time) or 2) Hardware optimization that reduces the latency of each IO operations from 6 s to 5 s.

Finally, how long would it take if we apply both optimizations?