The Good, The Bad and The Ugly of EDA - PowerPoint PPT Presentation

1 / 36
About This Presentation
Title:

The Good, The Bad and The Ugly of EDA

Description:

The Good, The Bad and The Ugly of EDA – PowerPoint PPT presentation

Number of Views:223
Avg rating:3.0/5.0
Slides: 37
Provided by: ToddL150
Category:

less

Transcript and Presenter's Notes

Title: The Good, The Bad and The Ugly of EDA


1
The Good, The Bad and The Ugly of EDA
  • ???? ??? ???

2
??? ToddLin_at_emc.com.tw
  • Experience
  • 2000-Now ????, ???
  • 1996-2000 General Manager of Asia Operations
  • Professional Services Group
  • Synopsys, Inc.
  • 1990-1995 Director, GE/Aerospace, USA
  • IEEE 802.3 Standard Committee
  • Education
  • 1992 Sloan School of Management,
    Massachusetts Institute of Technology, USA
  • 1986 ??????, ?????

3
Agenda
  • In those good old days
  • Disintegration -gt Missing Links
  • The size does matter
  • You get what you measure
  • Q A

4
Cast
  • The Ugly
  • Tuco (Eli Wallach)
  • The Good
  • Blondie (Clint Eastwood)
  • The Bad
  • Sentenza (Lee Van Cleef)

5
In those good old days,Ugly and Good ride side
by side
  • The Ugly
  • IC Designer
  • Architecture
  • Manner
  • Floor
  • Micro
  • The Good
  • CAD Engineer
  • Algorithm
  • Model
  • Flow
  • Macro

6
Disintegration
7
The Bad Design Quality Issues
8
Implementation is very
Mask
9
Sharpen the Tools
  • You will need a lot of tools and models along the
    design process.
  • Make them available and sharp enough before you
    need it.

10
EDA is critical to success but
  • EDA company can only provide point tools to
    fulfill partial common need.
  • IC designer may not understand how to use the
    complex tools correctly
  • Most of the IC house buy tools from all major EDA
    company worsen the situation
  • In house CAD does not have budget, man-power and
    RD capability to integrate a schedule-predictable
    design flow

11
Electronics Design Automation
  • CS job but requires in depth EE knowledge
  • Your customers hate you but can not live without
    you
  • Point tools are hard to survive

12
  • Large Scale Software

13
The Size Does Matter ...
14
Why does it take so long ...
A Programming System (Interface, Integration)
A Program
3 X
3 X
A Programming Product (Generalization,
Documentation, Maintenance)
A Programming Systems Product
15
Men and Months are not interchangeable
  • Good cooking takes time.
  • If you are made to wait, it is to serve you
    better, and to please you.
  • Luncheon Entrees
  • Your lunch entree includes a soup and salad of
    your choice.Soup Alligator, Vichyssoise, Bisque
    de crevettes, Onion Soup, Gumbo.Salad Salad
    Antoine with Thousand Island, or House Oil and
    Vinegar Dressing.
  • Desserts
  • Chocolate Mousse Caramel Custard Creme
    GlaceeCherries Jubilee Fresh Fruit CupMeringue
    Glacee Sauce Chocolat
  • Restaurant Antoine
  • 713 Rue St. Louis New Orleans, LA 70130
    504.581.4422 FAX 504.581.3003

16
Why Did The Tower of Babel Fail ?
17
Some tasks can not be hurried
  • Written Specification
  • Had better to make it executable
  • Formal Definition/Common Language
  • IEEE has done a lot of ground work for you
  • Multiple Implementations
  • How do you know it is a good design ? (Quality
    Assessment)

18
Aristocracy and System Design
  • Conceptual integrity is the most important
    consideration in system design.
  • Someone or a small group of agreeing minds must
    control the concepts. This is an aristocracy that
    needs no apology.

19
Surgical Team Organization
  • Each team member is highly professional in
    different operation
  • A chief can coordinate parallel operations with
    very limited communication
  • Good programmers may be ten times as productive
    as poor ones

20
Calling The Shot
  • To estimate is hard but because it is hard to
    make it great
  • 0.60.8K line of Assembly code per man-year for
    OS
  • 23K line of Assembly code per man-year for
    compiler
  • High level language may achieve as much as 5X

21
Hatching a Catastrophe
  • How does a project get to be a year late ?
  • one day at a time
  • How does the project get killed ?
  • suddenly
  • Milestone must be
  • concrete, specific, measurable events defined
    with knife-edge sharpness
  • Schedule slip is a morale killer
  • to push hard is essential for great programming
    team, just as for great baseball team

22
Change is the only constant
  • Plan the product architecture for change
  • Tier/Layer/Modular architecture
  • Plan the organization for change
  • Matrix operation
  • Also plan to throw one away
  • Due to the fact that we dont know what we dont
    know

23
2nd System Effect
  • ???? of the first system
  • Lost conceptual integrity
  • Too ambitious to integrate too much good stuff

24
10 Pounds in a 5-Pound Sack
25
Do this! Do that!And nothing is going to
happen.
26
(No Transcript)
27
A Lot of Missing Roles and Doc
28
Worlds Largest Undocumented Device
  • Investigations revealed that Stonehenge was built
    in several stages from 2800 - 1800 BC.
  • It seems to have been designed to allow for
    observation of astronomical phenomena - summer
    and winter solstices, eclipses, and more.

29
What is necessary document ?
30
System Requirements Analysis/ Design
System Requirements Analysis System Design Software Requirements Analysis Preliminary Design Detailed Design Coding and CSU Testing CSC Integration and Testing CSCI Testing System Integration and Testing
System Specification
Software Design Document(s) (Det. Design)
Source Code Listings
Software Design Documentation (Prel. Design)
Preliminary System Specification
System Segment Design Document
Source Code
Updated Source Code
Software Test Plan
Preliminary Software Requirement Specification(s)
Software Test Description(s) (Cases)
Software Test Description(s) (Procedures)
Software Requirements Specification(s)
Software Test Report(s)
Deliverable Products
Operation and Support Documents
Preliminary Interface Requirements Specification
Interface Design Document
Interface Requirements Specification(s)
Preliminary Interface Design Document
Version Description Document(s)
Software Development Plan
Developmental Configuration
Software Product Specification(s)
Reviews and Audits
System Requirements Review
System Design Review
Preliminary Design Review
Software Specification Review
Critical Design Review
CSCI Functional Physical Configuration Audits
Test Readiness Review
Functional Baseline
Allocated Baseline
Product Baseline
Baselines
31
Pivot Traceability Matrix
  • Requirements
  • Specification
  • Test Cases
  • Doc Index

32
  • You get what you measure

33
Quality A Magic or A Trick ?
  • For system engineer
  • Any one can define a spec, but is it a right spec
    or even a good spec ?
  • For software engineer
  • Any programmer can write programs, but do the
    programs work ?
  • For hardware engineer
  • Any manufacturing foundry can produce the HW, but
    are they tested enough ?

34
Intangible Capability Maturity
  • How capable is your organization of consistently
    delivering product?
  • How mature is your process such that it helps you
    to consistently meet your customers expectations?
  • It is not a canned process, it is a model that
    identifies process areas and practices that are
    important to achieving an organization with
    mature capability to consistently deliver to the
    customer.
  • It helps an organization set up process on a
    first things first basis.

35
Conclusion
  • Disintegration drive
  • Professionalism
  • Missing link
  • EDA, Product Engineering
  • Large scale software ?
  • Most of them are peopleware
  • Accumulate checklist
  • Make the auditing process automated
  • You have to measure Quality of your work
  • If you dont know where you are, a map would not
    help.

36
Education through Head Training through Hands
  • Questions Comments
Write a Comment
User Comments (0)
About PowerShow.com