Scribd
Upload
9 views19 pages

Ratio Games

The document discusses 'Ratio Games', which involve comparing systems using various metrics and ratios to determine performance. It includes case studies comparing processors, guidelines for using appropriate metrics, and highlights the potential misuses of percentages in performance comparisons. The document emphasizes the importance of selecting the right base for comparisons and provides exercises for practical application of the concepts discussed.

Uploaded by

hz7rshm6qb
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views19 pages

Ratio Games

The document discusses 'Ratio Games', which involve comparing systems using various metrics and ratios to determine performance. It includes case studies comparing processors, guidelines for using appropriate metrics, and highlights the potential misuses of percentages in performance comparisons. The document emphasizes the importance of selecting the right base for comparisons and provides exercises for practical application of the concepts discussed.

Uploaded by

hz7rshm6qb
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 19

Ratio Games

©2010 Raj Jain www.rajjain.com

11-1
Overview

‰ Ratio Game Examples


‰ Using an Appropriate Ratio Metric
‰ Using Relative Performance Enhancement
‰ Ratio Games with Percentages
‰ Ratio Games Guidelines
‰ Numerical Conditions for Ratio Games

©2010 Raj Jain www.rajjain.com

11-2
Case Study 11.1: 6502 vs. 8080

1. Ratio of Totals
‰ Conclusion: 6502 is worse. It takes 4.7% more time than 8080.

©2010 Raj Jain www.rajjain.com

11-3
6502 vs. 8080 (Cont)
2. 6502 as the base: 3. 8080 as the base:

1. Ratio of Totals: 6502 is worse.


It takes 4.7% more time than 8080.
2. With 6502 as a base: 6502 is better.
It takes 1% less time than 8080.
3. With 8080 as a base: 6502 is worse. It takes 6% more time.
©2010 Raj Jain www.rajjain.com

11-4
Case Study 11.2: RISC vs. CISC

‰ Conclusion: RISC-I has the largest code size. The second


processor Z8002 requires 9% less code than RISC-I.
©2010 Raj Jain www.rajjain.com

11-5
RISC vs. CISC (Cont)

‰ Conclusion: Z8002 has the largest code size and that it takes
18% more code than RISC-I. [Peterson and Sequin 1982]
©2010 Raj Jain www.rajjain.com

11-6
Using an Appropriate Ratio Metric
Example:

1. Throughput: A is better
2. Response Time: A is worse
3. Power: A is better
©2010 Raj Jain www.rajjain.com

11-7
Using Relative Performance Enhancement
‰ Example: Two floating point accelerators

‰ Problem: Incomparable bases. Need to try both on the


same machine

©2010 Raj Jain www.rajjain.com

11-8
Ratio Games with Percentages
‰ Example: Tests on two systems
System A:

System B:

1. System B is better on both systems


2. System A is better overall.

©2010 Raj Jain www.rajjain.com

11-9
Percentages (Cont)

‰ Other Misuses of Percentages:


¾ 1000% sounds more impressive than 11-time. Particularly if
the performance before and after the improvement are both
small
¾ Small sample sizes disguised in percentages
¾ Base = Initial. 400% reduction in prices ⇒ Base = Final
©2010 Raj Jain www.rajjain.com

11-10
Ratio Games Guidelines
1. If one system is better on all benchmarks, contradicting
conclusions can not be drawn by any ratio game technique

©2010 Raj Jain www.rajjain.com

11-11
Guidelines (cont)
2. Even if one system is better than the other on all benchmarks,
a better relative performance can be shown by selecting
appropriate base.
¾ In the previous example, System A is 40% better than
System B using raw data, 43% better using system A as a
base, and 42% better using System B as a base.
3. If a system is better on some benchmarks and worse on others,
contracting conclusions can be drawn in some cases. Not in all
cases.
4. If the performance metric is an LB metric, it is better to use
your system as the base
5. If the performance metric is an HB metric, it is better to use
your opponent as the base
6. Those benchmarks that perform better on your system should
be elongated and those that perform worse should be
shortened
©2010 Raj Jain www.rajjain.com

11-12
Numerical Conditions for Ratio Games
‰ Raw Data: ‰ A is better than B iff

‰ With A as the Base: ‰ A is better than B iff

©2010 Raj Jain www.rajjain.com

11-13
Numerical Conditions (Cont)
‰ With B as the base: ‰ A is better than B iff

©2010 Raj Jain www.rajjain.com

11-14
Numerical Conditions (Cont)

Ratio of B/A response on 2


B is better
benchmark j
using all 3
1
A is Base B
better
using all 3 Raw Data
Base A
0
11 12 3
Ratio of B/A response on
benchmark i

©2010 Raj Jain www.rajjain.com

11-15
Summary

‰ Ratio games arise from use of incomparable bases


‰ Ratios may be part of the metric
‰ Relative performance enhancements
‰ Percentages are ratios
‰ For HB metrics, it is better to use opponent as the base

©2010 Raj Jain www.rajjain.com

11-16
Exercise 11.1
‰ The following table shows execution times of three
benchmarks I, J, and K on three systems A, B, and C.
Use ratio game techniques to show the superiority of
various systems.

©2010 Raj Jain www.rajjain.com

11-17
Exercise 11.2
‰ Derive conditions necessary for you to be able to use
the technique of combined percentages to your
advantage.

©2010 Raj Jain www.rajjain.com

11-18
Homework
‰ Read chapter 11
‰ Submit answer to Exercise 11.1

©2010 Raj Jain www.rajjain.com

11-19

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