Compare and contrast pipeline to non-pipeline system design. Describe how pipelining improves the throughput and latency of the system. Illustrate with examples. Also include Hardware complexities, size, cost, and power consumption involved with the technology. Create an comparison chart with all points.
In the non-pipelined system design the instruction gets the access to memory and control structures immediately whereas if we look in pipelined system designs an instruction needs to wait for the previous instruction. So if we talk about a single instruction non pipeline is faster.For the pipelined system design we would need number of stages in pipeline to find out CPI for running multiple instructions.
The idea in pipelined processor is to split the processor instruction into series of steps or we can say independent stages and each stage is aligned to perform certain task. We can divide stages like fetch unit, decode unit, execute unit , write unit and two different instructions can run parallel as one might be at decode stage and other at fetch stage there by taking less time and increase throughput and performance.