Question

A nonpipelined system takes 300ns to process a task. The same task can be processed in a 5-segment pipeline with a clock cycle of 60ns. Determine the speedup ratio of the pipeline for 100 tasks

Answer 1

Pipelined architecture is 1.667 times more speedy than non pipelined architecture.

Step-by-step explanation:

There are five stages to to complete an instruction to execute in pipeline architecture. as

Stage 1: Instruction Fetch

Stage 2: Instruction Decode

Stage 3: Execute

Stage 4: Memory Access

Stage 5 : Write Back

The pipeline architecture processed instruction as given below. Each Stage take 1 clock cycle, which is represented as CC.

Stage 1 Stage 2 Stage 3 Stage 4 Stage 5

I1: 1cc 2cc 3cc 4cc 5cc

I2 2cc 3cc 4cc 5cc 6cc

I3 3cc 4cc 5cc 6cc 7cc

I4 4cc 5cc 6cc 7cc 8cc

each four Instructions set takes 8 Clock Cycles.

I5 9cc 10cc 11cc 12cc 13cc

I6 10cc 11cc 12cc 13cc 14cc

I7 11cc 12cc 13cc 14cc 15cc

I8 12cc 13cc 14cc 15cc 16cc

.

.

.

I97: 193cc 194cc 195cc 196cc 197cc

I98 194cc 195cc 196cc 197cc 198cc

I99 195cc 196cc 197cc 198cc 199cc

I100 196cc 197cc 198cc 199cc 200cc

Pipelined architecture Time Calculation

So in pipeline architecture 100 instructions takes 200 clock cycles to execute.

1 Clock Cycle time period is = 60ns.

200 Clock Cycle = 60ns * 200 = 12000 ns

Non Pipelined architecture Time Calculation

1 task Time period= 300ns

100 task time period = 100 * 300 = 30000 ns

The difference between pipelined and non pipelined architecture time period is = 30000 - 12000 = 18000 ns

Speedup ratio

Speed up ratio= 30000/18000 = 1.667