Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

Assume that individual stages of the data path have the following latencies :- I

ID: 3832487 • Letter: A

Question

Assume that individual stages of the data path have the following latencies :-

IF

ID

EX

MEM

WB

150ps

150ps

120ps

130ps

140ps

a. What is the clock cycle time in a pipelined and a single cycle non-pipelined processor?

b. What is the latency of an instruction in a pipelined and a single cycle non-pipelined processor?

c. What are the speed-up that could be achieved with the pipelined processor over the non- pipelined processor for two programs containing 10 instructions and 10000 instructions, respectively?

IF

ID

EX

MEM

WB

150ps

150ps

120ps

130ps

140ps

Explanation / Answer

a)Time for 1 instruction in pipeline = 150ps = max(150, 150, 120, 130, 140)

Clock cycle time in pipelined = 6.67 GHz

Time for 1 instruction in non-pipeline = 690ps = add(150, 150, 120, 130, 140)

Clock cycle time in Non- pipelined = 1.449 GHz

b) latency of an instruction in a pipelined = 150*5 = 750ps

latency of an instruction in a non-pipelined = 690ps

c) Time Require to execute 10 instructions:

Pipeline = 14*150 = 2100ps

Non-Pipeline = 690*10 = 6900 ps

SpeedUp = 6900/2100 = 3.2857

Time Require to execute 100 instructions:

Pipeline = 104*150 = 15600ps

Non-Pipeline = 690*100 = 69000 ps

SpeedUp = 69000/15600 = 4.423