For your research, you profiled an application that requires 1000 seconds of execution time on a single-core architecture. You would like to design a multi-core architecture to speed up this particular application. Use Amdahls Law to answer the following questions:

(a) If you were to rewrite the application to take advantage of a multi-core architecture, what would be the maximum achievable speedup on an architecture with four cores? What would the new execution time be? Assuming 900 seconds of the applications execution time is perfectly parallelizable, while 100 seconds must be executed sequentially. Also assume the performance of each core in the quad-core architecture equals that of the original single-core architecture.

(b) What would be the execution time and speedup achievable with an eight-core machine?

(c) What would be the maximum speedup achievable on a machine with unlimited cores?

(d) What would be the maximum speedup achievable on a machine with unlimited cores if 95% of the execution time could be perfectly parallelized?