New version page

BU EECE 352 - Computer Performance

Type: Lecture Note
Pages: 3

This preview shows page 1 out of 3 pages.

View Full Document
View Full Document

End of preview. Want to read all 3 pages?

Upload your study docs or become a GradeBuddy member to access this document.

View Full Document
Unformatted text preview:

EECE 352 1st Edition Lecture 4 Outline of Last Lecture I. Synchronous Circuit Design Techniques Outline of Current Lecture II. Computer PerformanceA. Defining PerformanceB. Relating The Metrics Current LecturePerformance Metrics:- Purchasing Perspective:-looking for combination of best performance/lowest cost- Design Perspective:-when faced with design options, which has the best performance improvement and lowest cost?**Both require basis for comparison and metric for evaluationI. Performance Criteria:A. CPU1. Instruction Set & Compiler2. Microarchitecture Implementation3. VLSI SpeedB. Memory and I/O1. Size2. SpeedC. OS Overhead1. Single User2. Time SharingDefining (Speed) Performance: The ultimate criterion: >Execution time of user program>The total amount of work done in a given time These notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.>Note: Decreasing response time almost always improves throughput**To maximize performance, need to minimize execution time… We define: Performance = Throughput/Response Time = Throughput/Execution Time Performance is proportional to (1/Execution Time)**How are response time and throughput affected by -Replacing the processor with a faster version? -Adding more processors? Want to distinguish elapsed time and the time spent on our task- Elapsed time- Total response time, including all aspects including processing, I/O, OS overhead, idle time -Determines system performance - CPU time: CPU execution time -Time the CPU spends working on a task -Does not include time waiting for I/O or running other programs -Comprises user CPU time and system CPU time -Different programs are affected differently by CPU and system performanceRelating the Metrics:- CPU execution time for a program = Number of CPU clock cycles for a program X Clock cycle timeor, CPU execution time for a program = Number of CPU clock cycles for a program Clock rate- Can improve performance by reducing either the length of the clock cycle or the number of clock cycles required for a program - Example: We have Computer A, CPU time = 10s, clock rate = 100MHz. We are asked to design Computer B, CPU time = 6s, what is the clock rate needed to make Computer B to take 1.2 times as many CPU clocks as Computer A ?**Clock Cycles per Instruction -Not all instructions take the same amount of time to execute…One way to think about execution time is that it equals the number of instructions executed multiplied by the average time per instructionNumber of CPU clock cycles for a program = Instruction Count * CPI; CPI = Average Clock Cycles per Instruction*Instruction count depends on ISA and the compiler…Weighted Average CPI: CPI = Clock Cycles/Instruction


View Full Document
Loading Unlocking...
Login

Join to view Computer Performance and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Computer Performance and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?