ese370: circuit-level modeling, design, and optimization ...ese370/fall2017/handouts/lec12.pdf ·...

ESE370: Circuit-Level Modeling, Design, and Optimization for Digital Systems

Lec 12: October 4, 2017 Scaling

Penn ESE 370 Fall 2017 - Khanna

!  VLSI Scaling Trends/Disciplines !  Effects !  Alternatives (cheating)

Penn ESE 370 Fall 2017 - Khanna 2

Scaling

!  Premise: features scale “uniformly” "  everything gets better in a predictable manner

!  Parameters: #  λ (lambda) -- Mead and Conway (Day14) #  F -- Half pitch – ITRS (F=2λ) #  S – scale factor – Rabaey

#  F’=S×F

ITRS Roadmap

!  International Technology Roadmap for Semiconductors "  Try to predict where industry going

!  ITRS 2.0 started in 2015 with new focus "  System Integration, Heterogeneous Integration,

Heterogeneous Components, Outside System Connectiviy, More Moore, Beyond CMOS and Factory Integartion.

!  http://www.itrs2.net/

Microprocessor Trans Count 1971-2015

5 Kenneth R. Laker, University of Pennsylvania, updated 20Jan15

Curve shows transistor count doubling every

two years Pentium

4004 8006

8080 Mot 6800

Mot 68000 80286

MOS 6502 Zilog Z80

AMD K5 Pentium II

Pentium III AMD K7

Pentium 4 AMD K8

AMD K10 AMD 6-Core Opteron 2400 4-Core i7

2-Core Itanium 2 6-Core i7 6-Core i7 16-Core SPARC T3

10-Core Xenon IBM 4-Core z196 IBM 8-Core POWER7

4-Core Itanium Tukwilla

  2015: Oracle SPARC M7, 20 nm CMOS, 32-Core, 10B 3-D FinFET transistors.

2015 Penn ESE 370 Fall 2017 - Khanna

Trend – “Minimum Feature Size vs. Year

Process Node/”Minimum” Feature

Year 1960 1980 2000 2020 2040

100 µm

10 µm

0.1 µm

0.1 nm

Integrated Circuit History

0.18 µm in 1999 ITRS Roadmap

Transition Region

Quantum Devices

Atomic Dimensions

“Minimum” Feature Measure = line/gate conductor width or half-pitch (adjacent 1st metal layer lines or adjacent transistor gates)

NOT SO Distant Future

Intel Cost Scaling

http://www.anandtech.com/show/8367/intels-14nm-technology-in-detail

Moore’s Law Impact on Intel uComputers

8 2010 YEAR

Serial data links operating at 10 Gbits/sec.

Increased reuse of logic IP, i.e. designs and cores.

2BT µP (Intel Itanium Tukwila) 4-Core chip (65 nm) introduced Q1 2010.

3BT mP (Intel Itanium Poulson) 8-Core chip (32 nm) to be introduced 2012.

Introduces 22 nm Tri-gate Transistor Tech.

Complexity - # transistors Double every Two Years 0.022um

0.032um 2009

Min Feature

More Moore $ Scaling

!  Geometrical Scaling "  continued shrinking of horizontal and vertical physical

feature sizes

!  Equivalent Scaling "  3-dimensional device structure improvements and new

materials that affect the electrical performance of the chip even if no geometrical scaling

!  Design Equivalent Scaling "  design technologies that enable high performance, low

power, high reliability, low cost, and high design productivity even if neither geometrical nor equivalent scaling can be used

22nm 3D FinFET Transistor

Tri-Gate transistors with multiple fins connected together

increases total drive strength for higher performance

http://download.intel.com/newsroom/kits/22nm/pdfs/22nm-Details_Presentation.pdf

High-k gate

dielectric

More-than-Moore

“More-than-Moore”, International Road Map (IRC) White Paper, 2011.

International Technology Road Map for Semiconductors

Semiconductor System Integration – More Than Moore's Law

Transistors/cm2

ponents/cm2

1970 1980 1990 2000 2010 2020

Multichip Module

System- in-package

(SIP) System-

on-package (SOP)

R. Tummala, “Moore's Law Meets Its Match”, IEEE Spectrum, June, 2006

SOP law for system integration. As components shrink and boards all but disappear, component density will double every year or so.

Improvement Trends for VLSI SoCs Enabled by Geometrical and Equivalent Scaling

!  TRENDS: !  Higher Integration level

"  exponentially increased number of components/transistors per chip/package.

!  Performance Scaling "  combination of Geometrical

(shrinking of dimensions) and Equivalent (innovation) Scaling.

!  System implementation "  SoC + increased use of SiP -

!  CONSEQUENCES: !  Higher Speed

"  CPU clock rate at multiple GHz + parallel processing.

!  Increased Compactness & less weight "  increasing

system integration.

!  Lower Power "  Decreasing energy

requirement per function.

!  Lower Cost "  Decreasing cost per

function.

Societal Needs

More Moore $ Scaling

!  Examples: "  Design-for-variability "  Low power design (sleep modes, clock gating, multi-

Vdd, etc.) "  Multi-core SOC architectures

Preclass 1

!  Scaling from 32nm $ 22nm? "  Scaling minimum gate length "  And pitch distance

Half Pitch (= Pitch/2) Definition

(Typical MPU/ASIC)

(Typical DRAM)

 Poly  Pitch

 Metal  Pitch

Source: 2001 ITRS - Exec. Summary, ORTC Figure, Andrew Kahng Penn ESE 370 Fall 2017 - Khanna 17

MOS Transistor Scaling - (1974 to present)

S=0.7 per technology node

[0.5x per 2 nodes] Pitch Gate

Source: 2001 ITRS - Exec. Summary, ORTC Figure, Andrew Kahng Penn ESE 370 Fall 2017 - Khanna 18

250 -> 180 -> 130 -> 90 -> 65 -> 45 -> 32 -> 22 -> 16

0.7x 0.7x

N N+1 N+2

Linear Time

1994 NTRS - .7x/3yrs

Actual - .7x/2yrs

19 Penn ESE 370 Fall 2017 - Khanna Source: 2001 ITRS - Exec. Summary, ORTC

Figure, Andrew Kahng

Node Cycle Time:

Scaling Calculator

Scaling

!  Channel Length (L) !  Channel Width (W) !  Oxide Thickness (Tox) !  Doping (Na) !  Voltage (V)

Full Scaling (Ideal Scaling)

!  Channel Length (L) S !  Channel Width (W) S !  Oxide Thickness (Tox) S !  Doping (Na) 1/S !  Voltage (V) S

Effects on Physical Properties and Specs?

!  Area !  Capacitance !  Resistance !  Threshold (Vth) !  Current (Id) !  Gate Delay (τgd) !  Wire Delay (τwire) !  Power

!  λ % λS ! Area impact? !  Α = L × W!  Α % ΑS2

!  32nm % 22nm !  50% area !  2 × transistor capacity

for same area

W S=0.7

Capacitance

!  Capacitance per unit area scaling?

"  Cox= εSiO2/Tox

"  Tox% S×Tox

"  Cox % Cox/S

Capacitance

!  Gate Capacitance scaling?

#  Cgate= A×Cox

#  Α % Α×S2

#  Cox % Cox/S#  Cgate % S×Cgate

Resistance

! Resistance scaling? ! R=ρL/(W*t) ! W$ S×W! L, t remain similar (not scaled) ! R $ R/S

Threshold Voltage

! VTH% S×VTH

Current

!  Which Voltages matters here? (Vgs,Vds,Vth…) !  Transistor charging looks like

voltage-controlled current source !  Saturation Current scaling?

Id=(µCOX/2)(W/L)(Vgs-VTH)2

Vgs=V$ S×V

VTH$ S×VTH

W$ S×WL$ S×LCox $ Cox/S

Current

Vgs=V$ S×V

VTH$ S×VTH

W$ S×WL$ S×LCox $ Cox/S

Id=(µCOX/2S)(SW/SL)(SVgs-SVTH)2

Current

Vgs=V$ S×V

VTH$ S×VTH

W$ S×WL$ S×LCox $ Cox/S Id$ S×Id

Current

!  Velocity Saturation Current scaling?

Vgs=V$ S×V

VTH$ S×VTH

L$ S×L W$ S×WCox $ Cox/S

Current

Vgs=V$ S×V

VTH$ S×VTH

VDSAT $ S×VDSAT

VDSAT ≈Lνsatµn

Current

Vgs=V$ S×V

VTH$ S×VTH

VDSAT $ S×VDSAT

Id$ S×Id

IDS ≈νsatCOXW VGS −VTH −VDSAT

VDSAT ≈Lνsatµn

Gate Delay

#  Gate Delay scaling? #  τgd=Q/I=(CV)/I #  V$ S×V

#  Id $ S×Id

#  C $ S×C

Note: Ids modeled as current source; V is changing with scale

factor

Gate Delay

#  Gate Delay scaling? #  τgd=Q/I=(CV)/I #  V$ S×V

#  Id $ S×Id

#  C $ S×C

#  τgd $ S×τgd

Note: Ids modeled as current source; V is changing with scale

factor

Wire Delay

#  Wire delay scaling? #  τwire=R×C

#  R $ R/S #  C $ S×C #  τwire $ τwire

!  …assuming (logical) wire lengths remain constant...

Power Dissipation (Dynamic)

!  Capacitive (Dis)charging scaling?

! P=(1/2)CV2f

! V$ S×V

! C $ S×C

! P$ S3×P

Power Dissipation (Dynamic)

!  Capacitive (Dis)charging scaling?

! P=(1/2)CV2f

! V$ S×V

! C $ S×C

! P$ S3×P

!  Increase Frequency?

!  τgd $ S×τgd

! So: f $ f/S

! P $ S2×P

Effects?

!  Area S2 !  Capacitance S !  Resistance 1/S !  Threshold (Vth) S !  Current (Id) S !  Gate Delay (τgd) S !  Wire Delay (τwire) 1 !  Power S3, S2 (w/ freq scaling)

Power Density

!  P% S2P (increased frequency)!  P% S3P (same frequency)!  A % S2A

!  Power Density: P/A two cases? "  P/A % P/A increase freq. "  P/A % S×P/A same freq.

Cheating…

!  Don’t like some of the implications !  High resistance wires !  Higher capacitance !  Atomic-scale dimensions

!  …. Quantum tunneling

!  Need for more wiring !  Not scale speed fast enough

Improving Resistance

! R=ρL/(W×t) ! W$ S×W! L, t similar ! R $ R/S

Improving Resistance

! R=ρL/(W×t) ! W$ S×W! L, t similar ! R $ R/S

What might we do? Didn’t scale t quite as fast $ now taller than wide.

Decrease ρ (copper) – introduced 1997 http://www.ibm.com/ibm100/us/en/icons/copperchip/

Capacitance and Leakage

!  Capacitance per unit area "  Cox= εSiO2

"  Tox% S×Tox

"  Cox % Cox/S

What’s wrong with tox = 1.2nm?

source: Borkar/Micro 2004

Capacitance and Leakage

!  Capacitance per unit area "  Cox= εSiO2

"  Tox% S×Tox

"  Cox % Cox/S

What might we do? Reduce dielectric constant, ε, and increase

thickness to mimic tox scaling. Penn ESE 370 Fall 2017 - Khanna 45

ITRS 2009 Table PIDS3B Low Operating Power Technology Requirements

Grey cells delineate one of two time periods: either before initial production ramp has started for ultra-thin body fully depleted (UTB FD) SOI or multi-gate

(MG) MOSFETs, or beyond when planar bulk or UTB FD MOSFETs have reached the limits of practical scaling (see the text and the table notes for further

discussion).

Year of Production 2009 2010 2011 2012 2013 2014 2017 2017 2017 2018 2019 2020 2021 2022 2023 2024 MPU/ASIC Metal 1 (M1) ½ Pitch (nm) (contacted) 54 45 38 32 27 24 21 18.9 16.9 15 13.4 11.9 10.6 9.5 8.4 7.5 Lg: Physical Lgate for High Performance logic (nm) 29 27 24 22 20 18 17 15.3 14 12.8 11.7 10.7 9.7 8.9 8.1 7.4 Lg: Physical Lgate for Low OperatingPower (LOP) logic (nm) [1] 32 29 27 24 22 18 17 15.3 14 12.8 11.7 10.7 9.7 8.9 8.1 7.4 EOT: Equivalent Oxide Thickness (nm) [2] Extended planar bulk 1 0.9 0.9 0.85 0.8 UTB FD 0.9 0.85 0.8 0.75 0.7 MG 0.8 0.8 0.75 0.73 0.7 0.7 0.65 0.65 0.6 0.6 Gate poly depletion (nm) [3] Bulk 0.27 0.27 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Channel doping (E18 /cm3) [4] Extended Planar Bulk 3 3.7 4.5 5 5.5 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Junction depth or body Thickness (nm) [5] Extended Planar Bulk (junction) 14 13 11.5 10 9 UTB FD (body) 7 6.2 6 5.1 4.7 MG (body) 8 7.6 7 6.4 5.8 5.4 4.8 4.4 4.2 4 EOTelec: Electrical Equivalent Oxide Thickness (nm) [6] Extended Planar Bulk 1.64 1.53 1.23 1.18 1.14 UTB FD 1.3 1.25 1.2 1.15 1.1 MG 1.2 1.2 1.15 1.13 1.1 1.1 1.05 1.05 1 1

High-K dielectric Survey

Wong/IBM J. of R&D, V46N2/3P133—168, 2002 Penn ESE 370 Fall 2017 - Khanna 47

Intel NYT Announcement

!  Intel Says Chips Will Run Faster, Using Less Power "  NYT 1/27/07, John Markov "  Claim: “most significant change in the

materials used to manufacture silicon chips since Intel pioneered the modern integrated-circuit transistor more than four decades ago”

"  “Intel’s advance was in part in finding a new insulator composed of an alloy of hafnium…will replace the use of silicon dioxide.”

Wire Layers = More Wiring

Gate Delay

#  τgd=Q/I=(CV)/I #  V$ S×V #  Id=(µCOX/2)(W/L)(Vgs-VTH)2

#  Id $ S×Id

#  C $ S×C

#  τgd $ S×τgd

How might we accelerate?

Improving Gate Delay More

#  τgd=Q/I=(CV)/I #  V$ V #  Id=(µCOX/2S)(SW/SL)(Vgs-VTH)2

#  Id $ Id/S#  C $ S×C

#  τgd $ S2×τgd

Don’t scale V!

How might we accelerate?

But… Power Dissipation (Dynamic)

!  Capacitive (Dis)charging

#  P=(1/2)CV2f #  V$ V#  C $ S×C #  P$ S×P

But… Power Dissipation (Dynamic)

!  Capacitive (Dis)charging

#  P=(1/2)CV2f #  V$ V#  C $ S×C #  P$ S×P

!  Increase Frequency? #  f $ f/S2 #  P $ P/S

If don’t scale V, power dissipation doesn’t scale down!

…And Power Density

!  P$ P/S (increase frequency)!  Α $ S2×Α!  What happens to power density?

…And Power Density

!  P$ P/S (increase frequency)!  Α $ S2×Α!  What happens to power density?

!  P/A $ (1/S3)P

!  Power Density Increases

…this is where some companies have gotten into trouble…

Historical Voltage Scaling

!  Frequency impact? !  Power Density impact?

http://software.intel.com/en-us/articles/gigascale-integration-challenges-and-opportunities/

Scale V separately with Factor U

!  τgd=Q/I=(CV)/I ! V$U×V

!  τgd=Q/I=(CV)/I ! V$U×V!  Id=(µCOX/2S)(SW/SL)(UVgs-UVTH)2

!  Id $ U2/S×Id

! C $ S×C

!  Id $ U2/S×Id

! C $ S×C

!  τgd $ (SU/(U2/S))×τgd

!  τgd $ (S2/U)×τgd

!  Id $ U2/S×Id

! C $ S×C

!  f $ (U/S2)×f

Ideal scale factors: S=1/100 U=1/100 τ=1/100 fideal=100

!  Id $ U2/S×Id

! C $ S×C

!  f $ (U/S2)×f

What are U and S?

!  Id $ U2/S×Id

! C $ S×C

!  f $ (U/S2)×f

Cheating factors: S=1/100 U=1/10

How much faster are gates?

!  Id $ U2/S×Id

! C $ S×C

!  f $ (U/S2)×f

Cheating factors: S=1/100 U=1/10 τ=1/1000 fcheat=1000 fcheat/fideal=10

Power Density Impact

!  P = 1/2CV2 f !  P $ S U2 (U/S2) = U3/S !  P/A = (U3/S) / S2 = U3/S3

!  U=1/10 S=1/100 !  P/A $ 1000 (P/A)

!  Compare with ideal scaling: !  P/A $ (1/S3)P (ideal scaling) !  P/A $ 1,000,000 (P/A) (ideal scaling)

uProc Clock Frequency

The Future of Computing Performance: Game Over or Next Level? National Academy Press, 2011

http://www.nap.edu/catalog.php?record_id=12980

uP Power Density

The Future of Computing Performance: Game Over or Next Level? National Academy Press, 2011

http://www.nap.edu/catalog.php?record_id=12980 Penn ESE 370 Fall 2017 - Khanna 68

Conventional Scaling

!  Ends in your lifetime !  Perhaps already:

"  "Basically, this is the end of scaling.” "  May 2005, Bernard Meyerson, V.P. and chief technologist for

IBM's systems and technology group

ITRS 2.0 Report 2015

!  “After 2021, the report forecasts, it will no longer be economically desirable for companies to continue traditional transistor miniaturization in microprocessors.”

BUT…

Source:https://newsroom.intel.com/newsroom/wp-content/uploads/sites/11/2017/09/mark-bohr-on-continuing-moores-law.pdf

BUT…

Source:https://newsroom.intel.com/newsroom/wp-content/uploads/sites/11/2017/09/mark-bohr-on-continuing-moores-law.pdf

Big Ideas

!  Moderately predictable VLSI Scaling "  unprecedented capacities/capability growth for

engineered systems "  change

"  be prepared to exploit "  account for in comparing across time "  …but not for much longer

!  HW5 "  More transistor practice "  Hard – prepares you for design project 1 "  Due Wednesday

!  Midterm "  Grades and solutions posted "  Pick up from me after class

ese370: circuit-level modeling, design, and optimization ...ese370/fall2017/handouts/lec12.pdf ·...

Documents

penn ese370 fall2013 -- dehon 1 ese370: circuit-level...

intel versus amd: 2 ghz northwood versus athlon xp...

penn ese370 fall 2011 -- townley & dehon ese370:...

tv wonder pro - welcome to amd | processors ... wonder pro...

engineering a cache-oblivious sorting algorithm · pentium...

hoo ÷ $ # sx2210 - ø - dell · ú ' õ ' ' $ # ' '÷ö...

03 empregabilidade - marcos haniu.ppt [modo …...01 micro...

sabiqal.files.wordpress.com · personal komputerjenis...

penn ese370 fall2014 -- dehon 1 ese370: circuit-level...

penn ese370 fall2011 -- dehon 1 ese370: circuit-level...

ese370: circuit-level modeling, design, and...

penn ese370 fall2013 -- dehon 1 ese370: circuit-level...

01 - histórico da computação -...

ese370: circuit-level modeling, design, and optimization

características de equipos. hp características componente...

Векторизация кода · intel sse3: 2004,...

arquitectura de procesadores (introducción) -...

future time traveller€¦ · web viewcpu con support sse2,...

penn ese370 fall2010 -- dehon 1 ese370: circuit-level...

penn ese370 fall2012 -- dehon 1 ese370: circuit-level...