1940433059 - ComputerOrganization CDA3103 Dr.HassanForoosh

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
    Computer Organization CDA 3103 Dr. Hassan Foroosh  Dept. of Computer Science UCF © Copyright Hassan Foroosh 2004
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Popular Views of Performance PC Magazine Oct 7, 1997 pp. 22-23 ad by SAG electronics: “SAG offers the best price/performance” attributed to Network Computing June 1, 1997 P. 48 Benchmark Tests: “These Pentium MMX/233 notebooks were great performers,  posting Business Winstone scores that averaged about 21% higher  than the average scores of 166 MHz notebooks in our August  roundup” p. 216 ad by DEC: “…you’ll grow to love running Windows NT ®  software on the  world’s fastest microprocessor” What do you think these statements mean? How would you determine whether they are true?
Background image of page 2
    Today’s Lecture: Performance Performance:  What is it? Why is it important? How do we measure it? The CPU performance equation Execution time as the measure What affects execution time Examples Popular alternative metrics Reasons they don’t work Choosing good benchmarks Choosing bad benchmarks Comparing and summarizing performance Good and bad practices Amdahl's Law
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Performance is  Time to Complete Work Plane Boeing 747 BAC/Sud Concorde Speed 610 MPH 1350 MPH DC to Paris 6.5 hours 3 hours Passengers 470 132 Throughput (PMPH) 286,700 178,200 Time to do the task  (Execution Time) execution time, response time, latency Tasks per unit time (sec, minute, . ..) throughput, bandwidth
Background image of page 4
    Time and Frequency Dimensions Clock Period (seconds) Time Unit Frequency Unit 1 second (s) Hertz (Hz) 10 -3 millisecond (ms) Kilohertz (KHz) 10 -6 microsecond ( μ s) Megahertz (MHz) 10 -9 nanosecond (ns) Gigahertz (GHz) 10 -12 picosecond (ps) Terahertz (THz)
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Performance as Response  Time Performance is most often measured as response  time or execution time for some task.  “X is  n  times faster than Y”  means Performance(X)  Execution Time(Y)  ––––––––––––––     =   ––––––––––––––––    =   n Performance(Y)  Execution Time(X) Example Execution time of program P  X is 5 sec  Y is 10 sec. X is 2 times faster than Y
Background image of page 6
    What Time to Measure? Elapsed time, wall-clock time Actual time from start to completion Depends on CPU, system, I/O, etc. Often used in real benchmarks Only suitable choice when I/O is included CPU Time Measure/analyze CPU performance only May be suitable when machine is timeshared Possibly both user and system component User CPU time is our focus for first part of course Elapsed time = CPU time + Idle time When time is accurately accounted for
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 39

1940433059 - ComputerOrganization CDA3103 Dr.HassanForoosh

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online