Scribd
Upload
11 views

Computer Performance

Hdjcjejf

Uploaded by

bouraslmhajbi
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
11 views

Computer Performance

Hdjcjejf

Uploaded by

bouraslmhajbi
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17

Introduction to Computer Organization

Computer Performance
Overview
• Performance evaluation
• Limitations
• Metrics
• Processor performance equation
• Performance evaluation reports
• Amdahl’s law
Performance Evaluation
Program

Compiler

ISA

Microarchitecture

Hardware

Manufacturing
Performance Evaluation Concepts
• Computer performance evaluation is based on:
– Throughput (bits per second)
– Response time (a.k.a. execution time, elapsed time)
• Component- and system-level performance
• Processor performance evaluation
– Based on execution time of a program:
PerformanceX = 1/ Execution timeX
Relative performance: n = PerformanceX / PerformanceY
n > 1 => X is n times faster than Y
• Terminology
– Improve performance = increase performance
– Improve execution time = decrease execution time
Example
• Impact on throughput and response time of:
– Using a faster processor
• Decrease in response time
• Increase in throughput
– Adding more processors to a system
• Increase in throughput
• Decrease in response time (if overhead is low)
– In Massively Parallel Processors (MPP)
– In Symmetric Multiprocessing Processors (SMP)
» Only if additional processors reduce queue time
Measuring Performance
Components of the execution time of a program:
1. CPU execution time
• User CPU time
• System CPU time
2. I/O time
3. Time spent running other programs
Unix time command 9.7u = 9.7 sec User CPU time
time cc prog.c 1.5s = 1.5 sec System CPU time
9.7u 1.5s 20 56% 20 = Total Elapsed Time
56% = Percent CPU time
CPU Performance Equation
• CPU time = CPU clock cycles * cycle time
= CPU clock cycles / clock rate
– CPU clock cycles = IC * CPI
• IC: instruction count (number of instructions per program)
• CPI: average cycles per instruction

CPU time = IC * CPI * cycle time


seconds instructions clock cycles seconds
= * *
program program instruction clock cycle
• CPU clock cycles = Σi (CPIi * ICi)
– ICi : count of instructions of class i
– CPIi : cycles that takes to execute instructions of class i
Scope of Performance Sources
CPU time = IC * CPI * Cycle time
Program

Compiler

ISA

Microarchitecture

Hardware

Abstraction level interdependence


Example 1
• A program runs in 10 secs on a 2.0 GHz processor.
A designer wants to build a new computer that can
run the program in 6 secs by increasing the clock
frequency. However the average “new” CPI will be
1.2 times higher.
• What faster clock rate should the designer use?

10 IC * CPI / 2 GHz (current exec. time)


=
6 IC * 1.2 CPI / X GHz (target execution time)
Solve for X, to obtain X = ___ GHz
Example 2
• Comparing two compiler code segments
Instruction class i CPIi for instruction class i
A 1
B 2
C 3

Code Instruction counts (ICi) for instruction class


sequence A B C
1 2 1 2
2 4 1 1

• Which code sequence executes the most instructions?


• Which will be faster? S1 = 2 . 1 + 1 . 2 + 2 . 3 = 10 cyc
S2 = 4 . 1 + 1 . 2 + 1 . 3 = 9 cyc
Components of the CPU Equation
• IC Instruction count
– Compiler
• Instruction set simulator
• Execution-based monitoring (profiling)
• CPI if pipelined execution is used
CPIi = Pipeline CPIi + Memory CPIi
• Clock cycle time
– Timing estimators or verifiers (complete design)
– Target cycle time
Performance Evaluation Programs
• Ideal situation: known programs (workload)
• Benchmarks
– Real programs
– Kernels
– Toy benchmarks
– Synthetic benchmarks
• Risk: adjust design to benchmark requirements
– (partial) solution: use real programs
• Engineering or scientific applications
• Software development tools
• Transaction processing
• Office applications
Performance Reports
• Reproducibility
– Include hardware / software configuration
– Evaluation process conditions
• Summarizing performance
– Total time: CPU time + I/O time + Other time
– Arithmetic mean: AM = 1/n * Σ exec timei
– Harmonic mean: HM = n / Σ (1/ratei)
– Weighted mean: WM = Σ wi * exec timei
– Geometric mean: GM = (Π exec time ratioi)1/n
GM (Xi) Xi
= Important Stuff
GM (Yi) Yi
Arithmetic and Geometric Means
A B C Execution time (in seconds)
machines: A, B, and C
P1 1 10 20
programs: P1 and P2
P2 1000 100 20
Amdahl’s Law
• Law of diminishing returns
Execution time before improvement
Speedup =
Execution time after improvement

Example: Two factors - F1: 75% improve, F2: 50% improve

F1
F2

1
Speedup =
(1 - fraction enhanced) + (fraction enhanced/factor of improvement)
Example
• A program runs in 10 seconds
• What is the speedup after a faster floating
point unit is incorporated?

FP Unit 5 times faster 5 seconds in FP operations


6 2
1

Speedup
Speedup = 1.9
5
(1 – Fraction FP) + FractionFP 1.8
4 5 1.7
1.6
3 1.5 1
Speedup =
1.4 0.5
2
1.3 0.5 + FP factor of improvement
1 1.2
Fraction of FP 1.1 FP Unit im provem ent
0 1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Conclusions

• Many different performance data


– CPU time, I/O time, Other time, Total time
• Select best presentation method
– Arithmetic mean for execution times
– Geometric mean for performance ratios
• Watch out for Amdahl’s Law
– Diminishing returns w/ improved performance
– Impacts of development

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy