Scribd
Upload
54 views

Module9 Design Quality Estimation

This document discusses techniques for estimating hardware and software design quality metrics. It describes the need for quality metrics to evaluate designs against constraints and explore alternative solutions. Some key hardware metrics discussed include die size, pin count, clock cycles, and slack. Software metrics include program size, data memory size, and execution time. The document outlines processor-specific and generic models for estimating these metrics at different stages of the hardware/software codesign process.

Uploaded by

tauseef qidwai
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
54 views

Module9 Design Quality Estimation

This document discusses techniques for estimating hardware and software design quality metrics. It describes the need for quality metrics to evaluate designs against constraints and explore alternative solutions. Some key hardware metrics discussed include die size, pin count, clock cycles, and slack. Software metrics include program size, data memory size, and execution time. The document outlines processor-specific and generic models for estimating these metrics at different stages of the hardware/software codesign process.

Uploaded by

tauseef qidwai
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 39

BITS Pilani

presentation
BITS Pilani Swapnil Patil
Pilani Campus
BITS Pilani
Pilani Campus

Hardware Software Codesign


BITS Pilani
Pilani Campus

Chapter 9 Design Quality


Estimation
Reference

Chapter 7 : Design Quality Estimation

BITS Pilani, Pilani Campus


Chapter 9. Design Quality
estimation
– Quality Metrics
– HW Estimation
– SW Estimation

BITS Pilani, Pilani Campus


BITS Pilani
Pilani Campus

9.1 Quality metrics


Quality metrics

Need of quality metrics:


1) To evaluate the design quality by comparing estimates
for specific design matric with constraint specified for
that matric.
2) Provides designer to explore more alterative solutions by
providing quick feedback

BITS Pilani, Pilani Campus


Design model

BITS Pilani, Pilani Campus


Accuracy

BITS Pilani, Pilani Campus


Fidelity

• Fidelity of implementation is defined as the percentage


of correctly predicted comparison between design
implementation models

BITS Pilani, Pilani Campus


Fidelity

BITS Pilani, Pilani Campus


Quality metrics

• Hardware cost metrics


• Software cost metrics
• Performance metrics
• Power consumption metrics
• Design for testability
• Design time
• Time to metrics
• Manufacturing cost
• Communication metrics

BITS Pilani, Pilani Campus


Hardware cost metrics

• Die size
• Packing cost
• Testing cost
• Prorated engineering design cost
• Pin count
• Manufacturing Technology used

BITS Pilani, Pilani Campus


Software cost metrics

• Program size: measure of size of memory required to


store compiled code
• Data memory size: measure of amount of memory
needed to store all data values created & manipulated
during program execution.
• Development cost
• Design time

BITS Pilani, Pilani Campus


Performance metrics

Computation metric :
• Clock cycle
• Control steps
• Execution time

Communication metric

BITS Pilani, Pilani Campus


Clock cycle effect

BITS Pilani, Pilani Campus


Pipelined implementation

BITS Pilani, Pilani Campus


Stage delay & throughput

BITS Pilani, Pilani Campus


Execution time

BITS Pilani, Pilani Campus


DFT (Debug feature for HW)

BITS Pilani, Pilani Campus


BITS Pilani
Pilani Campus

9.2 HW Estimation
Hardware estimation model

BITS Pilani, Pilani Campus


BITS Pilani, Pilani Campus
Clock cycle estimation

Manually provided by designer

BITS Pilani, Pilani Campus


Maximum operator delay
method

BITS Pilani, Pilani Campus


Slack

Slack = idle time

BITS Pilani, Pilani Campus


Control step estimation

BITS Pilani, Pilani Campus


Datapath estimation

BITS Pilani, Pilani Campus


Storage allocation

BITS Pilani, Pilani Campus


Pin Estimation

BITS Pilani, Pilani Campus


BITS Pilani
Pilani Campus

9.3 Software estimation


Software estimation model

BITS Pilani, Pilani Campus


Software estimation model

Processor specific model


Generic estimation model

BITS Pilani, Pilani Campus


Processor specific model

BITS Pilani, Pilani Campus


Generic model

BITS Pilani, Pilani Campus


BITS Pilani, Pilani Campus
Program execution time

Static estimation (Process path)


Dynamic estimation (Process path + upper/lower bounds)

BITS Pilani, Pilani Campus


Program memory size

• Directly proportional to no of instructions needed by


program
• Program B instruction set G , size of each instruction
instr_size(g)

BITS Pilani, Pilani Campus


Data memory size

• The data memory size is determined by examining the


data declaration in the functional specification

BITS Pilani, Pilani Campus

You might also like