Evaluating cycles per arithmetic operation (CPOp) and Cycles per assembly instru

Question

Evaluating cycles per arithmetic operation (CPOp) and Cycles per assembly instruction (CPI) for a vector-multiply add benchmark.

a) Total elapsed execution time was collected using two difference test conditions for a vector multiply-add benchmark code and is recorded in the table below. Evaluate and fill the values for the execution time per vector, execution time per vector element, and CPOP for both test cases. Clock rate for the system is 2.4GHz.

b) evaluate the speed up (performance improvement) provided when using test condition 2 versus test condition 1

Speed up = ________

Test condition # vector elements, N # accuracy improvement loops Measured total execution time, sec Evaluated execution time per vector, N Evaluated execution time per vector element CPOp 1 1000 1E7 32.0 2 1000 1E7 3.0

Explanation / Answer

Solution:

the clock rate is 2.4 Ghz, execution time is 32 and 3

CPOP of test 1= (32)/(2.4 * 10^9)= 0.00001333333*10^7= 0.133

CPOP of test 2= (3)/(2.4 * 10^9) * 10^7=  0.0125

b)

Speedup= 0.133/0.0125= 10.64?

I hope this helps if you find any problem. Please comment below. Don't forget to give a thumbs up if you liked it. :)

Test condition # vector elements, N # accuracy improvement loops Measured total execution time, sec Evaluated execution time per vector, N Evaluated execution time per vector element CPOp 1 1000 1E7 32.0 32.0 32/1000= 0.032 2 1000 1E7 3.0 3.0 3/1000= 0.003

Related Questions

Navigate

• Browse (All)
• Browse E
• Subjects

---

---

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