Pipelined vs Non-Pipelined Processors

 Directions: It can be helpful to examine in detail how pipelining affects the underlying clock cycle time of the processor. We will assume that the individual stages of the datapath have specific latencies: IF – 250ps ID – 350ps EX – 150ps MEM – 300ps WB – 200ps We’ll assume that the instructions executed by the processor are broken down as follows: ALU/logic – 45% Jump/branch – 20% LDUR – 20% STUR – 15% Using the information above, in a Word document, answering the following questions: 1) Calculate the clock cycle time in a pipelined processor. 2) Calculate the clock cycle time in a non-pipelined processor. 3) What is the total latency of an LDUR instruction in a pipelined and non-pipelined processor? 4) If we can split one stage of the pipelined datapath into two new stages, each with half of the latency of the original stage, which stage would you split and why? What would be the new clock cycle time of the processor? 5) Assuming there are no stalls or hazards, what is the utilization of the data memory? 6) Assuming there are no stalls or hazards, what would be the utilization of the write-register port of the “Registers” unit?

#Pipelined #NonPipelined #Processors

Share This Post

Email
WhatsApp
Facebook
Twitter
LinkedIn
Pinterest
Reddit

Order a Similar Paper and get 15% Discount on your First Order

Related Questions