Suppose that the system has some performance problem. You are thinking that if you could boost overall system performance by 12% (making it 12% faster than it is now), you could eliminate the problem. One option is to use a faster processor. If the processor accounts for 25% of the workload of this system, and you need to boost performance by 12%, how much faster does the new processor need to be?

Suppose that the system has some performance problem. You are thinking that if you could boost overall system performance by 12% (making it 12% faster than it is now), you could eliminate the problem. One option is to use a faster processor. If the processor accounts for 25% of the workload of this system, and you need to boost performance by 12%, how much faster does the new processor need to be?

Systems Architecture

7th Edition

ISBN:9781305080195

Author:Stephen D. Burd

Publisher:Stephen D. Burd

Chapter4: Processor Technology And Architecture

Section: Chapter Questions

Problem 2PE: If a microprocessor has a cycle time of 0.5 nanoseconds, what’s the processor clock rate? If the...

See similar textbooks

Similar questions

On the other hand, it has been suggested to use a processor that can decode encrypted data and system instructions, and then execute the instructions on the decoded data and system instructions. The CPU will then encrypt the suits that have been provided. What are the advantages of using a CPU of this caliber? What are the architectural requirements for a processor of this nature?
Alice designs a 2 GHz processor where two important programs, A and B, take one second each to execute. Each program has a CPI of (1.5). Elena is tasked with designing the company's next-generation processor. She comes up with an idea that improves the CPI of A to 1.1 and the CPI of B to 1.4. But the idea is so complex that the processor can only be implemented with a cycle time of 0.6 ns. Does Elena's new processor out-perform Alice's processor on program A? How about on program B?
a program runs in 10 seconds on computer A, which has a 4 GHz clock. We are trying to help a computer designer build a computer, B, which will run this program in 6 seconds. The designer has determined that a substantial increase in the clock rate is possible, but this increase will affect the rest of the CPU design, causing computer B to require 1.2 times as many clock cycles as computer A for this program. What clock rate should we tell the designer to target?
A program runs in 15 seconds on computer X, which has a 2 Ghz clock. You are required to help a computer designer to build another computer, Y, that will run the program in 10 seconds. The designer has determined that a substantial increase in the clock rate is possible, but this increase will affect the rest of the CPU design, causing computer Y to require 1.5 times as many clock cycles as computer for this program. What clock rate should you tell the designer to target? Select one: a.0.3 Ghz b.3 Ghz c.9 Ghz d.8 Ghz
Our favorite program runs in 10 seconds on computer A, which has a 2 GHz clock. We are trying to help a computer designer build a computer, B, which will run this program in 6 seconds. Th e designer has determined that a substantial increase in the clock rate is possible, but this increase will affect the rest of the CPU design, causing computer B to require 1.2 times as many clock cycles as computer A for this program. What clock rate should we tell the designer to target?
Our favorite program runs in 8 seconds on computer A, which has a 2 GHz clock. We are trying to help a computer designer build a computer, B, that will run this program in 4 seconds. The designer has determined that a substantial increase in the clock rate is possible, but this increase will affect the rest of the CPU design, causing computer B to require 1.3 times as many clock cycles as computer A for this program. What clock rate should we tell the designer to target?
Suppose we have 10 tasks for CPU1 to perform, and CPU1 takes 3 units of time to perform a task. o What is the throughput per unit of time? o Suppose we add 3 processors to our system which are identical to CPU1 so we have a total of 4. What would the new throughput per unit of time be for the same 10 tasks? What is the new response time?
When it comes to process execution, what is the advantage of first allocating it to the I/O queue and then pushing it to the CPU when it is ready? What, in your opinion, will happen if the I/O is not completed correctly? Is the CPU burst going to be affected? What do you mean?
System A has two processors, and Program X takes 10 seconds to run on one of them. Program Y executes in parallel on the other CPU in 10 seconds. System B is powered by a single CPU that can only handle one problem at a time. On this CPU, Program X takes 6 seconds to execute. Which system would you choose if overall system throughput was important to you?
How can we prevent a CPU from wasting cycles by processing several instructions at once? Batch-process those instructions. Could you tell me how much this plan's practicality depends on particular components? Please describe one of their effects.
When feasible, what are the upsides of prioritizing the I/O queue as the place where processes would be executed? If the I/O connection suddenly goes down, what do you expect to happen? How would this affect the fastest possible CPU burst rate? I'm not quite sure I get what you're getting at here.
Let's pretend that every 18 months a new generation of CPUs comes out that has twice as many cores as the last. How much more off-chip memory bandwidth will be needed for a CPU to be released in three years if per-core performance is to remain unchanged?