I\' am having a lot of trouble with this question so if you could include an exp

Question

I' am having a lot of trouble with this question so if you could include an explantion too that would be amazing.

Q3) Suppose that add, load, store, float_add, other instructions account for the following percentages, respectively, of all the execution time: 17%, 23%, 15%, 15%, 30%.

a) Hypothetically, if we are able to infinitely improve all instructions but one, we still only attain a speedup of 5.8823. What is the instruction that we couldn’t improve?

b) Due to an upgrade in the hardware, those instructions are speeded up by a factor of: 4, 3, 2, 5, 2, respectively. Using Amdahl’s law, compute the overall speedup.

Explanation / Answer

There are three possible ways to improve performance.

1. an enhanced compiler can reduce the floating-point instructions to ½ of the original floating-point instructions with the cost of increasing the integer instructions by 15% of the original integer instructions.

2.new pipeline technique can reduce the average CPI of the floating-point instruction from 10 to 4 with an increasing clock cycle time of 4%. Third, an improved caching technique reduces the CPI of Load/Store from 4 to 3

b. let Speedup =new rate /old rate

=old latency /new latency

Latency = Instruction Count * Cycles/Instruction * Seconds/Cycle

L = IC * CPI * CT

case1:

1÷((10.05)+(0.05)÷1.1)
=1.00456621

case 2:

1÷((10.04)+(0.04)÷1.1)=1.0045

case 3:

1÷((10.02)+(0.02)÷1.1)=1.0045

Related Questions

Navigate

• Browse (All)
• Browse I
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.