Introduction / Agenda

Bob Pfahl, iNEMITown & Country Hotel and

Convention CenterSan Diego, CA

October 9, 2009

1

iNEMI Technical Committee/PEG Chair Emulator Kick-off

Friday, October 9

TIME: TOPIC: DISCUSSION LEADER:8:00 AM Introduction and Review of Agenda: Bob Pfahl8:10 AM 2011 PEG Review Introduction Chuck Richardson8:20 AM Automotive PEG Review Discussion* Spall / All 8:55 AM Office PEG Review Discussion Lober / Aghazadeh 9:30 AM Break9:40 AM Network, Datacom, Telecom PEG Review John Duffy / Xue

10:15 AM Medical PEG Review Discussion TBD / All10:50 AM Consumer PEG Review Discussion Shahidzadeh/Roesch11:25 AM Aerospace/Defense PEG Review Murphy / Gieskes12:00 PM ITRS/iNEMI Coordination Juan-Antonio Carballo12:30 PM Lunch

*Each PEG presentation will consist of a 20 minute presentation by the PEG Chair followed by a 15 minute discussion led by the discussion leader. The assigned TC discussion leader should review the PEG’s material and presentation prior to the meeting.

22

iNEMI Technical Committee Business Mtg

Friday, October 9TIME: TOPIC: DISCUSSION LEADER:

1:00 PM T.C. Business Meeting1:00 PM Review 2009 TC Goals Jie Xue (Action)1:10 PM Project & Initiative Status Peter Tomaiuolo (Action)1:40 PM Presentation of 2+ SOW/PS from Asia Bob Pfahl (Action)1:50 PM International Activities & Recruiting Bob Pfahl (Information-Action)2:10 PM Discussion on new initiatives-workshops Bob Pfahl (Action)2:20 PM Research Committee-Research Priorities Update Alan Rae (Info)2:35 PM Leadership Discussion (Chairs etc.) Jie Xue (Action)2:50 PM 2010 calendar – Next Meeting Jie Xue (Wrap-up/Discussion)3:00 PM Adjourn

2010-11 T.C./ PEG Review

Introduction

Chuck RichardsonSMTAI, San Diego, CA

10/09/09

44

2011 iNEMI Roadmap Goals• Maintain/expand strong linkages with other technology roadmaps/organizations • Strengthen Product Emulator value:

– Utilize TWG Chair input on Key Attribute Spreadsheet formats prior to PEG kick-off– Leadership in place by SMTAI PEG kick-off meeting– Improve PEG Chapter availability for TWG use

• Strengthen linkages with European and Asian organizations • Expand emphasis on disruptive events (business and technical)• Expand emphasis on prioritizing technical and market gaps and needs throughout

roadmap• Improve the “Executive Summaries” in individual chapters by highlighting the key

points from the chapter• Establish Aerospace / Defense PEG for 2011 cycle • Transition TWG Chapter on Sensors to MEMS focus • Move organic substrates to “Packaging” TWG to reduce confusion • Restore TWG Chapter on Energy Storage & Conversion Systems • Utilize Web based meeting option to increase roadmap meeting efficiency

5

2010-11 Product Emulator Groups

Product Emulator Chair Potential PEG ParticipantsAutomotive Products

Jim Spall, Delphi

Delphi, Bosch, Continental

Consumer / Portable Products Shahrokh Shahidzadeh, Intel 3M, Nokia, Motorola

Defense/Aerospace Products TBD, Lockheed Martin Lockheed Martin, Boeing, Rockwell, Raytheon, BAE

Medical Products Anthony Primavera, Microsystems Engineering

Intel, Boston Scientific, Biotronik, TI

Office/Large Business System Products

David Lober, Intel Intel, IBM, HP, Sun

Telecom/Datacom Products TBD Alcatel-Lucent, Cisco, Huawei

66

2009 TWG Leadership Status

Business Processes / Technologies Chair(s) Co-Chair(s) Information Management Eric Simmon, NIST Jeff Pettinato, Intel Design Technologies Modeling, Simulation & Design Tools Yishao Lai, ASE S.B. Park, Binghamton U. Environmentally Conscious Electronics Need Chair Thermal Management Ravi Prasher, Intel Azmat Malik, Consultant

Manufacturing Technologies Final Assembly John Allen, Celestica Reijo Tuokko, Tampere U. Board Assembly Dongkai Shangguan,

Flextronics Aaron Unterborn, Microsoft Ravi Bhatkal, Cookson

Test, Inspection & Measurement Mike Reagin, Delphi

77

2011 TWG Leadership (cont.)Component / Subsystem Technologies

Chair(s) Co-Chair(s)

Semiconductor Technology Paolo Gargini, Intel Alan K. Allan, Intel

Optoelectronics Dick Otte, Promex William Ring, WSR

Photovoltaics Jim Handy, Objective-Analysis Alain Harrus, Cross Link

Packaging Bill Bottoms, NanoNexus William Chen, ASE

Passive Components Philip Lessner, Kemet

Connectors John MacWilliams, Consultant

RF Components Ken Harvey, Advantest Eric Strid, Cascade MicroTech

MEMS / Sensors John McKillop, Tekton LLC Raffaella Borzi, IMEC

Large Area, Flexible Electronics Dan Gamota, Printovate Jan Obrzut, NIST Jie Zhang, Printovate

Energy Storage & Conversion Systems

Need Chair Need Co-chair

Interconnect Substrates (Ceramic) Howard Imhof, Metalor Ton Schless, Sibco

Interconnect PCB John T. Fisher, IPC Henry Utsunomiya, Consultant

Mass Data Storage Roger F. Hoyt, Consultant Tom Coughlin, Coughlin Associates

Solid State Illumination Marc Chason, Consultant

88

2011 Roadmap Schedule

• 3Q2009: Recruit Product Sector Champions, teams and refine data charts/Begin 2011 Roadmap Newsletter & send 2009 PEG chapters 8/1/09

• 3/4Q09: Product Sector Champions Develop Emulators – September 9, 2009 – Teleconference with P.E. Group Chairs– September 17, 2009 Web based meeting TWG/PEG Chairs (key attributes)– October 9, 2009 - Roadmap PEG Kick-off with PEG/TWG/TC at SMTAI– December 17 , 2009 – Web based Teleconference with TC on PEG Emulator review

• 2009 “Word” chapter, format, Exec. Summary emailed to each TWG chair (Word) 1/4/2010

• Organizing Teleconference with TWG Chairs 1/11/2010: • February , 2010 PEG Workshop/TWG Kick-off CA:

– Product Sector Tables Complete – PEG Chapter rough drafts written– Cross cut issues are initially addressed

99

2011 Roadmap Schedule - Continued

• April 9, 2010 TC/PEG/TWG face to face chapter status review meeting at APEX

• May 6, 2010 Telecon with TWG Chairs, Preliminary PEG Chapters Due• May 2010 – Open Roadmap TWG Presentations in Las Vegas, NV (ECTC)• June , 2010 European Roadmap Workshop – TBD• June , 2010 – Asian Roadmap Workshop – TBD • July 1, 2010 – TWG Drafts Due for TC Review• August 4, 2010 – TC Face-to-Face Review with TWG Chairs at TBD• September 22, 2010 Final Chapters of Roadmap Due• October , 2010 Council of Members Briefing SMTAI • October 31, 2010 – Edit, Prepare Appendix A-D, Executive Summary • November 20, 2010 – Go To “Press” • December 5, 2010 – Ship to Members • April, 2011 – Industry presentation at APEX

2011 Product Emulator Descriptions

Emulator Characteristics

Consumer / Portable Produced in high volumes, cost is the primary driver, hand held battery powered products are also driven by size and weight reduction

Office / Large Business Driven by the need for maximum performance over a wide range of cost targets

Automotive Products Products that must operate in an automotive environment

Netcom Systems Products that serve the networking, datacom and telecom markets and cover a wide range of cost and performance targets

Medical Products Products that must operate with high reliability and, in some cases, support life critical applications

Aerospace / Defense Products that must operate reliably in extreme environments

10

11

Format for Product Emulator Chapters • Executive Summary (half page)• Introduction• Situation Analysis

– Benchmark state of Industry and Technology– Key Drivers: cost, performance, size, market– Chart 2009 vs. 2011 forecast differences

• Roadmap of Quantified Key Attribute Needs (2011 – 2021)• Critical (Infrastructure) Issues –

– Identify Paradigm Shifts – Provide Vision of Final Assembly Process– Discuss System Test

• Prioritized Technology Requirements and Trends: Research, Development, Implementation

• Contributors

12

PEG Chair Presentation Format:20 Minutes

• Emulator Spreadsheet Presentation– Suggested Critical New Rows / Columns for 2011– Summary of Changes to Existing Rows / Columns

• Preliminary Areas of Focus for 2011 Chapter– Business Issues– Technical Issues

13

T.C. Emulator Spreadsheet Review, Discussion Format: 15 Minutes

General Areas of Review, Feedback and Discussion:• Data / Format Review Summary

– Feedback/Discussion on PEG Suggestions Concerning New Parameters/Metrics/Format

– Any Situation Analysis Feedback• State of Industry Segment and Technology• Key Drivers: cost, performance, size, market?

– Any Critical (Infrastructure) Issues Feedback – Summary of Discussion / Suggested Changes

2010-11 T.C./ PEG Review –

Automotive

Jim Spall, DelphiSMTAI, San Diego, CA

10/09/09

2010-11 T.C./ PEG Review –Office/Large

Business Systems

David Lober SMTAI, San Diego, CA

10/09/09

16

Office/Large Business Systems PEG

•Office/Large Business System /Communication PEG includes:

–Computers and Office Equipment•Computers: Desktop and Notebook PCs, Workstations, and Servers•Computer Peripherals: Printers, Scanners, Input/Output Devices…•Storage Systems: Hard disk drives, SAN/NAS…•Office Equipment: photocopiers, fax machines, digital projectors…

17

Office/Large Business Systems PEG

• Preliminary Areas of Focus for 2011 Chapter– Business Issues

• Continuing decline of ASPs• Impact of low system price points (ie. Netbook/top)• Mobile systems volume surpassing desktop systems• Energy Cost• Environmental Concerns• Mega Data Centers & Cloud Computing

18

Office/Large Business Systems PEG

• Preliminary Areas of Focus for 2011 Chapter– Technical Issues

• Battery life• Power Management• Thermal management• Smaller form factors (x,y,z)• Meeting “Green” requirements at neutral cost• Number and speed of I/O continues to scale• Continued system integration (SOCs, SiPs, MCPs, MCMs..)

19

Office/Large Business Systems PEG

• Emulator Spreadsheet Presentation– Summary of Changes to Existing Rows / Columns

• Display: Ave Power Dissipation• Reliability: Humidity Range• Memory: Max Power, Min Speed• Power: Battery (Max Temp, Cost, Recharge time, Run time)• PCB Fab geos

20

Discussion & Emulator Feedback

General Areas of Review, Feedback and Discussion:• Require more participation in the PEG for success

– Need OEM participation for the key areas (Server, Notebook, & Desktop systems, storage, digital projectors…)

• Data / Format Review Summary– Feedback/Discussion on PEG Suggestions Concerning New

Spreadsheet Changes – Any Situation Analysis Feedback

• State of Industry Segment and Technology• Key Drivers: cost, performance, size, market?

– Any Critical (Infrastructure) Issues Feedback – Summary of Discussion / Suggested Changes

2010-11 T.C./ PEG Review –

Netcom

John DuffySMTAI, San Diego, CA

10/09/09

22

NetCom PEG Status Issues

• Committee status– A new chairperson needs to be identified– Strongly suggest representation from wireless telecom

segment– Continue representation from netcom & telecom companies– Suggest additional international representation, with a

balance between Europe, Asia, North America and developing regions (South America, Africa, Middle East, Southeast Asia-Oceania)

23

NetCom PEG Status Issues (2)

• Market Segment Status– Need to identify and detail areas of telecom / datacom /

netcom expansion (technology & geography)

• Areas for Chapter Concentration– Wireless netcom access (voice & data combined) is becoming

ubiquitous via new handheld 3G/4G cell products (iPhone & Blackberry)

– Network backbone bandwidth continues to expand with 40Gb/channel fiber optic deployment and investigation into 100Gb and 200Gb technologies

– WiMAX and SmartGRID are still primarily lab systems and have not yet started any wide deployment

24

Netcom PEG Segments

Parameter Segment Key Considerations

PWB Concentrate on high end / complex boards

Assembly Costs Is the Final Product a whole chassis or a 1U box?

Package Costs

Business Costs

Cycle Time Gather data for Product Production Life

Reliability Gather data for Vibrational Environment

Devices Compare with ASIC/Processor roadmaps

Passive Components

RF Components Focus on Cell / Wi-Fi / WiMax infrastructure technologies?

Display

25

Netcom PEG Segments

Parameter Segment Key Considerations

Memory Consider memory other than Flash & SRAM?

Components / Package

PWB / Substrates

Interconnects

Electrical Gather data for DPMO for Components

Power

Environmental Gather data for Compliance Dates for RoHS & WEEE

Thermal Consider the impact of liquid cooling systems?

Supply Chain Locations

Modeling Tools

2010-11 T.C./PEG Review –

Medical

Tony Primavera, Microsystems Engineering

SMTAI, San Diego, CA10/09/09

27

Introduction

• For the purposes of the Medical TIG Technical Plan and the Medical PEG, the product types are too varied to define global trends within the medical sector. Therefore, medical products will be grouped into three general categories. These have been separated in the PEG spread sheet.

– 1) Implanted products (those devices implanted in a human body)

– 2) Portable products (those devices that are easily transported)

– 3) Diagnostic imaging devices and large scale equipment, e.g., Ultrasound, MR, etc.

– Some product solutions will necessarily consist of combinations of all three categories of devices.

Medical Electronics Market

$0

$20

$40

$60

$80

$100$Bn

GEOGRAPHIC DISTRIBUTION

Americas

$66Bn5% of

Electronics Industry

2007

4.1% CAAGR2007-2013

$84Bn

$60Bn

$66Bn

54%

Japan

8%

Europe

25%

Asia/ROW

1 3%

0%

10%

20%

30%

40%

50%

60%

70%

2006 2007 2008 2009 2010 2011 2012 2013

ls108.273mw-med revPrismark

Overall market is approximately 70B$ (Non-IT), growing at a predicted 4.1% per year

Note: Tony’s opinion is this is still too low a predicted growth rate.

Prismark

29

Executive Summary

• Medical products are a varied sector with 3 diverging needs.

• 1. Consumer based medical products –– Require competitive cost and performance.

– Need to be available locally at the source of use.

– Relate closely to “Portable Sector” in terms of form factor, reliability and technical drivers.

– Primary region of assembly – ASIA, Europe and Americas.

– Primary area of Design: - US, Europe, ASIA.

– Emerging Market and manufacturing – India, Africa, South America.

– Primarily 2 to 8 layer PCBs. Include some high volume low layer count PCBs and flex for glucose monitors etc.

30

Executive Summary • 2. Large Scale Diagnostic equipment

– Driven by imaging and therapeutic requirements.

– Often large form factor PCBs, high density 12+ layer (can reach over 40 layers).

– Small components (01005’s and 0201’s) and large BGAs and connectors, are driving technology.

– Gaps in connectors and routing density exist today.

– End use – typically in fixed location (hospitals etc).

– Primary region of assembly – Americas, Europe, ASIA.

– Primary area of Design: - US, Europe.

– Emerging Market and manufacturing, especially in portable versions of the equipment (China, India).

31

Executive Summary • 3. Implantable / Life Critical Devices

– Primarily drive by therapy changes and reliability needs.

– Full traceability and validation of all processes is most often required.

– Space, circuit size and increased functionality are coupled with design to use less energy for many devices as they operate with a primary non-rechargeable battery. Technology driven by miniaturization.

– Often small form factor PCBs, high density 1 to 6 layer. Often utilize flex or HDI substrates. “Exotic” materials are more common place (PI, BT, Cyanate Ester, LCP, EPTFE, Ceramic and others).

– Typically utilize precious metals and involve soldering, adhesive processes and welding). Weld and final construction materials need to be bio-compatible, which primarily contain Titanium, Niobium, Platinum-Iridium and gold.

– Small components (01005’s and 0201’s) and high density stacked die or multichip BGAs are driving density needs. Reliability is primary “cost” driver.

– End use differs from location of sales (Patient versus hospital).

– Primary region of assembly – US, Ireland, Germany, Italy, France, Austria.

– Primary area of Design: - US, Europe.

– Emerging market and manufacturing, sub components (SE-ASIA).

32

Sector Overview – Portable Medical• Situation Analysis

– Growing market due to fundamental change in medical treatment. Dramatic change from reactive (prescribed physician treatment) to proactive and self (patient) involvement. Patient is more involved in their healthcare prevention and treatment.

– Key Drivers: cost, ease of use and minimal invasiveness of device or therapy, size. Many devices used world wide. Some very high volume (monitoring systems). Over-the counter and internet availability is high, which drives cost down.

– Growth areas: Largest growth in India and China.

– Challenges: • Technology: Battery rechargeable technology.• MFG: Manufacturing at point of use locations. Overall quality concern with supply chain

consistency.• Cost: Commodity status / global cost pressure. • Growing Concerns: RF crosstalk and interference with other devices.• Business Concerns: Counterfeit components and RoHS supply continuity• PCB Technology: Mixed. Most 2-6 layer, some HDI. Mostly FR4 derivatives

33

Sector Overview – Large Scale Medical• Situation Analysis

– Market growth overall slow due to global economy slowdown, hospital insolvency and reduction in capital spending by healthcare providers.

– Growing market in developing nations. Driving low cost / de-featured and portable equipment for rural areas. Key investments in China and India by large equipment OEMs and EMS providers for this infrastructure.

– Key Drivers: Image quality, moving video and live imaging, shrinking form-factor to make equipment portable. Reduction in cost. Region of use manufacturing and even design.

– Challenges: • Technology: Connector density, PCB density for HDI, massive use of small passives, and

heat generation/cooling needs.• MFG: Manufacturing at point of use locations. Overall supply chain consistency.• Cost: Commodity status / global cost pressure. • Growing Concerns: Connectors and board density. Offshore PCB manufacturing.• Business Concerns: Hospital Insolvency, lack of capital spending. Health care reform.• PCB Technology: Mixed. Mostly 12-14 layer, some over 40 layer HDI. Mostly FR4

derivatives

34

Sector Overview – Implantable• Situation Analysis

– Market growth overall increasing (globally estimated >10% YoY)

– Growing market in China, India, South America. Growth limited by 3 main factors: 1 lack of implanting physicians and facilities in many regions. 2. Lack of health care coverage / reimbursement. 3. Long development cycles and regulatory.

– Key Drivers: Reliability. Miniaturization. Increasing battery life for primary cells (ICDs for example). Increasing rechargeable longevity (Neuro-stimulators for example).

– Challenges: • Technology: PCBs – use of HDI and flex substrates. Lack of high voltage capacitors

(>1000V rating in smaller case sizes). • MFG: Maverick lot elimination. Lack of component quality standards for medical industry.

Overall supply chain consistency.• Cost: Suppliers exiting medical market, drive cost up. Developing nation cost pressure. • Growing Concerns: Supply base shrinking (especially in IC packaging area).• Business Concerns: Hospital Insolvency. Health care reform. Long reimbursement time.• PCB Technology: Mixed. 1 to 6 layer HDI. Materials run the gamut. FR4, BT, Polyimide,

Ceramic, Rigid-flex.

35

Sector Overview • Overall – still growing and stable market, with long term growth driven by aging

population and increased availability to medical care in rural areas. Supply chain concerns and lack of formal medical component reliability standards.

• PEG – rows missing / changes suggested?

– PCBs: 1 layer flex is missing (glucose strips for example)

– PCBS: suggest changing from FR4 and state-of the art in technology column to input of actual substrate used.

– Reliability: Pressure environment missing (hyperbaric chamber and scuba testing) –implantables

– Reliability: MRI safe / compatibility

– Reliability: bends/folds mechanical robustness

– Business: How to capture regional design / manufacturing for point of use applications.

– Business: Capture supply chain leaders somehow in component / PCB area. Maybe a survey of players in each respective market (ex. who makes 1000+ Volt caps? Or what companies make HDI flex? – Can be part of PEG, TIG or TWG?

2010-11 T.C./ PEG Review –

Portable & Consumer

Shahrokh Shahidzadeh, Intel

SMTAI, San Diego, CA10/09/09

Quick Consumer-Portable PEG Update

• September 17th meeting review resulted in few edits and additions to the 2009 baseline (see xls as needed)

– Some ideas on line item additions and specific chapter enhancements on RF and general mixed RF technologies, Environmental-Test, Corrosion , Memory Speed/Power/FF: DDR-eDRAM-NVRAM and thermal cycling discussed

– Also some items we never got to last cycle which need to consult with Susan on why/why not we should pursue

• Team membership and assignments and shortfalls will be addressed by end of Oct

– Have got one interested member from freescale on packaging (Ding Min), (need more referrals so appreciate team’s help)

• Plans in place to stay in synch with ITRS to cross collaborate

• Need to plan Medical and Automotive Exchange as there is overlap

Microsoft Excel Worksheet

In Preparation Of Feb’10 Workshop-TWG Kick Off

Team BiweeklyKick off (copy ITRS Optional)

PrismarkPlanningMeeting?

Product Sector Tables CompletePEG Chapter drafts writtenCross cut issues are initially address

1:1 Susan-Shahrokh

ITRS Synch2Juan Antonio& Team Phone/F2F

ITRS Synch1Juan AntonioPhone/F2F

Team Plan Mailer1. R&R2. Milestones3. Help Needed/more team members

Bi Weekly Team Meeting-Check PointsDeliverables: Industry scope-change

Drivers, Design Parameters & Segmentation, System Level Impact

(power, cost, perf, FF),

Auto+MedPEGs- Exchange

2010-11 T.C./PEG Review - Defense

and Aerospace

Bill Murphy, Lockheed MartinSMTAI San Diego, CA

October 09, 2009

Introduction

• Major thrust continues to be best value through use of COTS products• Low military volumes provide no leverage to influence component

suppliers relative to packaging design• As such, we are a user … not driver of new technology

– Although concerned about costs, cost will not be the driver– Bring a different set of constraints and thermal management approaches– Bring a unique set of applications and hence concerns

• Customers’ requirements mixed– Some still require 20 – 30 year service life in a harsh environment– Environmental concerns: temperature cycling, mechanical shock,

vibration, high altitude, and humidity– Logistic concerns: traceability, configuration management, and

obsolescence, counterfeit parts elimination, trust

Board Assembly• Low volume / high mix

• Unique process requirements– Conformal coating

– Adhesive bonding of PWB to heatsink frame

– Unique “ruggedization” requirements

– Many cleaning and material curing steps

– Attachment of odd form and solvent / heat sensitive components

– “SMT line” is small portion of total assembly process

• Reliability requirements drive additional and unique controls

• Unique traceability / configuration requirements– Unique controls to prevent unintended intrusion of Pb free content

• Additional ESS and customer acceptance test requirements– Fault isolation / test diagnostics

Reliability• Temperature Cycling

– “Rule of thumb” remains 1000 cycles of -55 to 125°C

– Some applications more severe, some less

• Vibration– Endurance and performance requirements

– Module level resonance frequency constraints driven by application

• Altitude – 15,000 feet to 70,000 feet to near orbit or deep space

• Mechanical Shock• Humidity

– 10 to 100% RH (Condensing)

• Salt / Fog

Passive Components

• Discrete passives going from 0201 to 01005

• Embedded passives– Defense industry is slow to implements (trend is to stay with proven

design techniques)

– Looking for improved design software tools

Components / Package• Gull wing constant - majority at 0.4mm pitch • Area Array

– Column Grid Array and Ball Grid Array– Wire Bond and Flip Chip– Variety of “heatsink” configurations

• Some drive cleaning and conformal coating concerns• Area Array Pitch

– Currently a mix from 1.27mm to 0.65mm pitch – Variety of CTEs dependent on die size and construction results in CTE

management headaches– Biggest Concerns are PWB wiring, thermal management, and environmental

• Obsolescence and logistics concerns • Leaded (SnPb) part availability / control

Interconnect PCB• Variety of materials

– FR-4, Polyimide, Aramid, “specialty”, and Duroid• Alternate finishes to SnPb without “black pad” risks• Use of “filled” vias

• CTE benefits of Aramid type materials required but continue to humidity concerns

• High I/O, sub 1.0mm pitch BGAs driving significant wiring challenges (many with strict wiring rules and impedance requirements)– Thicker PWBs, smaller holes, finer lines and spaces

– Module level constraints restrict PWB thickness

• Electrical requirements impedance requirements

• Continued demand for Rigid Flex type products although still a niche market

Electronic Connectors

• Mezzanine Connectors

• Compliant pin connectors

Environmental

• Most defense contractors sticking with eutectic tin lead solder– Pure tin finished component concerns and risk mitigation

requirements

– Pb free solder alloy concerns • Reballing, etc

– Efforts required to assess risks and control risks associated Pb free finishes and solders and required efforts to prevent unintended intrusions of Pb free products• Internal • Supply chain

Thermal Management

• Majority of PWA’s are conduction cooled

• Both component power and module power increasing significantly

• Will require significant thermal management materials and designs

Summary

• PWB materials

• PWB wiring density

• Unique card level assembly issues

• Product test complexity and diagnostic capabilities

• Unique reliability requirements

• Thermal management

• Mezzanine connectors

• Optoelectronics

• RF components and subsystems

International Technology

Roadmap for Semiconductors

Juan-Antonio Carballo, ITRS Design ITWG

SMTAI, San Diego, CA10/09/09

ITRS AND iNEMI

iNEMI Meeting, San Diego

Worldwide Design TWG

Presented by Juan-Antonio Carballo

ITRS-iNEMI Domain Space

Chip level System level

Techrequirements

Marketrequirements

iNEMI(emulators)

ITRS(Drivers)

ITRS Design Key Messages

DESIGN

- Updated design cost chart (design innovations in next 15 yrs)- Continued emphasis on SW/ system level design productivity

- DFM: new reliability roadmap, and "sigmas that DFM tools can reduce“- MPU / RAM models – design contribution to ORTC - More than Moore status: ("SiP alternative") - 2010

SYSTEM DRIVERS

- Updating key system (market) drivers in 2009 (consumer)- Final decision not to continue to broaden System Drivers

- New chair, introduction and new line of work for iNEMI collaboration - "SIP fabric driver" adding in 2009, point to TSV / other roadmaps as well- Refresh link with interc/A&P/PIDS – sensitivity analysis (portable) - 2010

Overview (2004-8)1. Increasingly quantitative roadmap2. Increasingly complete driver set

2004

2005

2006

2007

ExploreDesign metrics

Design Technology metrics

Revised Design metrics

Revised Design Technology Metrics

ConsumerPortableDriver

Consumer Stationary, PortableDrivers

Consumer Stationary,Portable,Networking Drivers

More Than Moore analysis + iNEMI

Driver study

System DriversChapter

DesignChapter

2008

Revised Design MetricsDFM extension

Updated Consumer Stationary,Portable,and Networking Drivers

More Than Moore extension+ iNEMI+ SW !!

2009

AdditionalDesign MetricsDFM ExtensionSystem level extension

Updated Consumer Stationary,Portable architecture,and Networking Drivers

More Than Moore extension+ iNEMI synch+ SW !!

ITRS Cost Chart 2009 rev1

(in Millions of Dollars)

$21 $16 $21 $21 $31 $24 $33 $15 $22 $16 $20 $19 $26 $33 $45 $29 $40 $25 $33 $27 $37$17 $22

$2 $8 $12 $18 $9 $13$20

$24$39

$30$41 $56

$79$34

$47$31

$42$27

$35 $34$47

$21 $29

$0

$50

$100

$150

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

Total HW Engineering Costs + EDA Tool Costs Total SW Engineering Costs + ESDA Tool Costs

IC Im

plem

enta

tion

Tool

Set

RTL

Func

tiona

l Ver

if. T

ool S

uite

Tran

sact

ion

Leve

l Mod

elin

g

Very

larg

e bl

ock

reus

e

AM

P Pa

ralle

l Pro

cess

ing

Inte

llige

nt T

estb

ench

Man

y C

ore

Dev

el. T

ools

SMP

Para

llel P

roce

ssin

g

Exec

utab

le S

peci

ficat

ion

Tran

sact

iona

l Mem

ory

Syst

em D

esig

n A

utom

atio

n

Design Impact on Power(Emphasis on System Level [SW/HW])

Alternative I

ITRS Drivers Space

Network ConsumerPortable

OfficeMedical Automotive Consumerstationary

Defense

MPU

PE(DSP)

AMS

Memory

Fabrics

Markets

Process For Each System Driver

Identify market requirements

Identify key design parameters

Select DRIVERrequirements

Create model Generate data

Color data

1

2 3

Select Critical/difficult parameters

PowerAreaHrs. operation

#unitsSize per unitMemoryPins

Power

Identify design requirements

Cost Perf.

#unitsSize of each unitNumber of pins

Need 2 synch with iNEMI emulatorsIndustry data

SOC Consumer Portable Chip Architecture Model (updated)

MainMemory

PE-1

Peripherals

PE-2 PE-n…

MainPrc.

MainPrc.

MainPrc.

MainPrc.

Function A Function B Function C

Function D Function E

MainMemory PE

PE

PE

PE

PE

MainPrc.

PE

PE

PE

PE

PE

PE

PE

PE

PE Peripherals

MainPrc.

MainPrc.

MainPrc.

《#Main Processor becomes more than 2 ,and will increase》

Complexity !

Figure SYSD5 SOC Consumer Portable Design Complexity Trends

28 36 45 59 77 98 127 166210

273347

435

560

725

889

1172

0

5

10

15

20

25

30

35

40

45

50

2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022

Logic, Memory Size (Norm

alized to 2007)

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2,000

# of Processing Engines

N um ber of Processing Engines(Right Y A xis)

Total Logic Size(N orm alized to 2007, Left Y Axis)

Total M em ory Size(Norm alized to 2007, Left Y A xis)

Power !

Figure SYSD6 SOC Consumer Portable Power Consumption Trends

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022

Power [m

W]

Trend: M em ory Static Pow er Trend: Logic Static Pow er

Trend: M em ory D ynam ic Pow er Trend: Logic D ynam ic Pow er

Requirem ent: D ynam ic plus Static Pow er

1st Alignment Between Chip and System RoadmapsConsumer portable pilot, focused on power/energy

~10 parameters to be aligned

iNEMI-ITRS Power ReconciliationTop-10 Parameters to be Reconciled

iNEMI VS ITRS (System V. Chip)Power Parameters’ Comparison

Voltage supply trends Power trends

iNEMI portable emulator

ITRS consumer driver

ITRS consumer driver

iNEMI portable emulator

ITRS stuck between lower voltages and higher power trends

PRIORITY: Future iNEMI-ITRSITRS Portable System Model

Application processor

Baseband processor

MemoryNAND Flash

MemoryWireless

Flash

Audio / video codec Power mgt.

I/O controller I/O transceivers

Oth

er (M

EM

S, e

tc.)

ProcessingPOWER

Memory / FlashCOST

Analog / I/ONOISE SENSITIVITY

Consumer Driver - Stationary

IO-M

emory IF &

Chip-to-C

hip IF -

Main

Processor

DPE

DPE

DPE

DPE

DPE

DPE

DPE

DPE

Main

Processor

DPE

DPE

DPE

DPE

DPE

DPE

DPE

DPE

Architecture Template

• SOC die size stays constant around 220mm2• Pair of a Main Processor and a number of DPEs structures basicarchitecture• Possible largest number of DPEswill be on this SOC to achieverequired performance• Circuit size of both Main Processorand DPEs stys constant• A Main Processor is assumed to be ableto control up to 8 DPEs

Design Trend: # of Processors & Processing Performance

5 8 8 13 1624 29

39 4761

77

99

120

149

189

248

73

2.0

10.1

0.8

0.256

0

50

100

150

200

250

3002005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

Year of Production

# of Components

0.1

1

10

100

Max Processing Power

[TFLOPS]

# of Main CPUs # of DPEs Max Processing Performance

Max

Pro

cess

ing

Perf

orm

ance

[TFL

OPS

]

Design Trend: Power Consumption – SOC Total

0

100

200

300

400

500

6002005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

Power Consumption [W]

Switching Power, Logic Switching Power, Memory

Leakage Power, Logic Leakage Power, Memory

600W

SOC total power consumption rapidly increases

More than Moore

Design-Driven Semiconductor Innovation (Moore Versus “More Than Moore”)

Domain / market independentInventory 50+ Design Solutions

Domain / market dependent(Consumer) driver architecture

48% (overall)

30% (# cores)

14-17% (Device speed)

SCALINGConventionalEquivalentDiversified

Moore+ Parameter Inventory

Classification of 50+ design technology solutions1. Supporting Moore’s Law – More Moore (geo scaling)2. Extending Moore’s Law – More Moore (equivalent scaling)3. Beyond Moore’s Law – More Than Moore (architecture diversification)

Example: System-Level Design Solutions

Design-Driven Semiconductor Innovation (Moore+)

Domain / market independent Inventory 50+ Design Solutions

Reveals impact of More Than Moore on key design phases

ITRS Design Key Messages

DESIGN

- Updated design cost chart (design innovations in next 15 yrs)- Continued emphasis on SW/ system level design productivity

- DFM: new reliability roadmap, and "sigmas that DFM tools can reduce“- MPU / RAM models – design contribution to ORTC - More than Moore status: ("SiP alternative") - 2010

SYSTEM DRIVERS

- Updating key system (market) drivers in 2009 (consumer)- Final decision not to continue to broaden System Drivers

- New chair, introduction and new line of work for iNEMI collaboration - "SIP fabric driver" adding in 2009, point to TSV / other roadmaps as well- Refresh link with interc/A&P/PIDS – sensitivity analysis (portable) - 2010

76

www.inemi.orgEmail contacts:

Chuck Richardson crichardson@inemi.org

Bob Pfahlbob.pfahl@inemi.org