Transcript ppt - SEAS

ESE370:
Circuit-Level
Modeling, Design, and Optimization
for Digital Systems
Day 13: September 27, 2013
Variation
1
Penn ESE370 Fall2013 -- DeHon
Previously
• Understand how to model
transistor behavior
• Given that we know its parameters
– Vdd, Vth, tOX, COX, W, L, NA …
CGC
CGCS
CGCB
2
Penn ESE370 Fall2013 -- DeHon
But…
• We don’t know its parameters (perfectly)
1.Fabrication parameters have error range
2.Identically drawn devices differ
3.Parameters change with environment
(e.g. Temperature)
4.Parameters change with time (aging)
Why I am more concerned with
robustness than precision.
3
Penn ESE370 Fall2013 -- DeHon
Today
• Sources of Variation
– Fabrication
– Operation
– Aging
• Coping with Variation
– Margin
– Corners
– Binning
4
Penn ESE370 Fall2013 -- DeHon
Fabrication
5
Penn ESE370 Fall2013 -- DeHon
Variation Types
• Many reasons why things are different
– Show up in many different ways.
• Scales
– Wafer-to-wafer, die-to-die, transistor-totransistor
• Correlations
– Systematic, spatial, random (uncorrelated)
6
Penn ESE370 Fall2013 -- DeHon
Source: Noel Menezes, Intel ISPD2007
7
Penn ESE370 Fall2013 -- DeHon
Process Shift
•
•
•
•
Oxide thickness
Doping level
Layer alignment
Growth and Etch rates and times
– Depend on chemical concentrations
• How precisely can we control those?
• Vary machine-to-machine, day-to-day
• Impact all transistors on wafer
8
Penn ESE370 Fall2013 -- DeHon
Systematic Spatial
• Parameters change consistently across
wafer or chip based on location
• Chemical-Mechanical Polishing (CMP)
– Dishing
• Lens distortion
9
Penn ESE370 Fall2013 -- DeHon
FPGA Systematic Variation
• 65nm
• Virtex 5
Penn ESE370 Fall2013 -- DeHon
[Tuan et al. / ISQED 2010]
10
Oxide Thickness
[Asenov et al. TRED 2002]
11
Penn ESE370 Fall2013 -- DeHon
Line Edge Roughness
• 1.2mm and
2.4mm lines
From:
http://www.microtechweb.com/2d/lw_pict.htm
12
Penn ESE370 Fall2013 -- DeHon
Optical Sources
• What is the shortest wavelength of
visible light?
• How compare to 45nm feature size?
13
Penn ESE370 Fall2013 -- DeHon
Phase Shift Masking
Today’s chips use λ=193nm
Source
http://www.synopsys.com/Tools/Manufacturing/MaskSynthesis/PSMCreate/Pages/default.aspx
14
Penn ESE370 Fall2013 -- DeHon
Line Edges (PSM)
Source:
http://www.solid-state.com/display_article/122066/5/none/none/Feat/Developments-in-materials-for-157nm-photoresists
15
Penn ESE370 Fall2013 -- DeHon
Intel 65nm SRAM (PSM)
Source:
http://www.intel.com/technology/itj/2008/v12i2/5-design/figures/Figure_5_lg.gif
16
Penn ESE370 Fall2013 -- DeHon
Statistical
Dopant
Placement
17
Penn ESE370 Fall2013 -- DeHon
[Bernstein et al, IBM JRD 2006]
Random Trans-to-Trans
•
•
•
•
Random dopant fluctuation
Local oxide variation
Line edge roughness
Etch and growth rates
– Stochastic process
• Transistors differ from each other in
random ways
Penn ESE370 Fall2013 -- DeHon
18
Source: Noel Menezes, Intel ISPD2007
19
Penn ESE370 Fall2013 -- DeHon
Impact
• Changes parameters
– W, L, tOX, Vth
• Change transistor behavior
– W?
– L?
– tOX?
IDS
IDS

VDSAT 
  satCOX W VGS  VT 


2 
2 
W 
VDS
 mnCOX  VGS VT VDS 

 L 
2 
20
Penn ESE370 Fall2013 -- DeHon
Example: Vth
• Many physical effects impact Vth
– Doping, dimensions, roughness
• Behavior highly dependent on Vth
IDS
IDS

VDSAT 
  satCOX W VGS  VT 


2 
W 
 IS e
 L 
Penn ESE370 Fall2013 -- DeHon
VGS VT 


 nkT / q 
 VDS 

kT / q 
1  e  1 VDS 


21
Vth Variability @ 65nm
22
Penn ESE370 Fall2013 -- DeHon
[Bernstein et al, IBM JRD 2006]
Impact of Vth Variation?
• Higher VTH?
– Not drive as strongly
– Id,vsat (Vgs-VTH)
– Performance?
23
Penn ESE370 Fall2013 -- DeHon
Impact Performance
• Vth  Ids  Delay (Ron * Cload)
24
Penn ESE370 Fall2013 -- DeHon
Impact of Vth Variation
Penn ESE370 Fall2013 -- DeHon
Think NMOS Vgs = Vdd
25
FPGA Logic Variation
• Xilinx Virtex 5
• 65nm
• Altera
Cyclone-II
• 90nm
[Tuan et al. / ISQED 2010]
Penn ESE370 Fall2013 -- DeHon
[Wong, FPT2007]
26
Variation in 65nm FPGAs
[Gojman, FPGA2013]
DeHon May 2013
27
LUT-to-LUT Same LAB
• LAB (27,22)
[Gojman, FPGA2013]
DeHon May 2013
average 5% variation
28
Delay Map
• LAB (27,22)
[Gojman, FPGA2013]
DeHon May 2013
29
Two LUT2LUT across Chip
30
DeHon May 2013
Reduce Vdd
(Cyclone IV 60nm LP)
[Gojman, FPGA2013]
DeHon May 2013
31
Impact of Vth Variation?
• Lower VTH?
– Not turn off as well  leaks more
IDS
W 
 IS e
 L 
Penn ESE370 Fall2013 -- DeHon
VGS VT 


 nkT / q 
 VDS 

kT / q 
1  e  1 VDS 


32
2004
Borkar (Intel) Micro 37 (2004)
Penn ESE370 Fall2013 -- DeHon
33
Operation
Temperature
Voltage
34
Penn ESE370 Fall2013 -- DeHon
Temperature Changes
• Different ambient environments
– January in Maine
– July in Philly
– Air conditioned machine room
• Self heat from activity of chip
• Quality of heat sink (attachment thereof)
IDS
W 
 IS e
 L 
Penn ESE370 Fall2013 -- DeHon
VGS VT 


 nkT / q 
 VDS 

kT / q 
1  e  1 VDS 


35
Self Heating
Borkar (Intel) Micro 37 (2004)
36
Penn ESE370 Fall2013 -- DeHon
Thermal Profile for Processor
37
Penn ESE370 Fall2013 -- DeHon
[Reda/IEEE Tr Emerging CAS v1n2 2011]
How does temperature impact
on-current?
• High temperature
– More free thermal energy
• Easier to conduct
• Lowers Vth
– Increase rate of collision
• Lower saturation velocity
• Lower saturation voltage
• Lower peak Ids  slows down
• One reason don’t want chips to run hot
38
Penn ESE370 Fall2013 -- DeHon
Temperature and Ids
39
Penn ESE370 Fall2013 -- DeHon
How does temperature impact
leakage current?
• High temperature Lowers Vth
IDS
W 
 IS e
 L 
Penn ESE370 Fall2013 -- DeHon
VGS VT 


 nkT / q 
 VDS 

kT / q 
1  e  1 VDS 


40
Voltage
• Power supply isn’t perfect
• Differs from design to design
– Board to board?
– How precise is regulator?
• IR-drop in distribution
• Bounce with current spikes
41
Penn ESE370 Fall2013 -- DeHon
Aging
Hot Carrier
NBTI
42
Penn ESE370 Fall2013 -- DeHon
Hot Carriers
• Trap electrons in oxide
– Also shifts Vth
43
Penn ESE370 Fall2013 -- DeHon
NBTI
• Negative Bias Temperature Instability
– Interface traps, Holes
• Long-term negative gate-source voltage
– Affects PFET most
• Increase Vth
• Partially recoverable?
• Temperature dependent
Another reason not to run hot.
Penn ESE370 Fall2013 -- DeHon
[Stott, FPGA2010]
44
Measured Accelerated Aging
(Cyclone III, 65nm FPGA)
Penn ESE370 Fall2013 -- DeHon
[Stott, FPGA2010]
45
Coping with Variation
46
Penn ESE370 Fall2013 -- DeHon
Variation
• See a range of parameters
– L: Lmin – Lmax
– Vth: Vth,min – Vth,max
47
Penn ESE370 Fall2013 -- DeHon
Impact of Vth Variation
• Higher VTH
– Not drive as strongly
– Id,vsat (Vgs-VTH)
• Lower VTH
IDS

VDSAT 
  satCOX W VGS  VT 


2 
– Not turn off as well  leaks more
W 
IDS IS e
 L 
Penn ESE370 Fall2013 -- DeHon
VGS VT 


 nkT / q 
 VDS 

kT / q 
1  e  1 VDS 


48
Variation
• Margin for expected variation
• Must assume Vth can be any value in range
– Speed  assume Vth slowest value
Ion,min=Ion(Vth,max)
Probability Distribution
Id,vsat (Vgs-Vth)
VTH
49
Penn ESE370 Fall2013 -- DeHon
Gaussian Distribution
From: http://en.wikipedia.org/wiki/File:Standard_deviation_diagram.svg
50
Penn ESE370 Fall2013 -- DeHon
Impact
• Given
– Vth,nom = 250mV
– Sigma 25mV
• Probability of 100 transistor circuit in
range when each has 96% prob. ?
• …when each has 99.8% probability?
51
Penn ESE370 Fall2013 -- DeHon
Impact
• Given
– Vth,nom = 250mV
– Sigma 25mV
• What maximum Vth should expect to
see for a circuit of
– 100 transistors?
– 1000 transistors?
– 109 transistors?
52
Penn ESE370 Fall2013 -- DeHon
Variation
• See a range of parameters
– L: Lmin – Lmax
– Vth: Vth,min – Vth,max
• Validate design at extremes
– Work for both Vth,min and Vth,max ?
– Design for worst-case scenario
53
Penn ESE370 Fall2013 -- DeHon
Margining
• Also margin for
– Temperature
– Voltage
– Aging: end-of-life
54
Penn ESE370 Fall2013 -- DeHon
Process Corners
• Many effects independent
• Many parameters
• With N parameters,
– Look only at extreme ends (low, high)
– How many cases?
• Try to identify the {worst,best} set of
parameters
– Slow corner of design space, fast corner
• Use corners to bracket behavior
Penn ESE370 Fall2013 -- DeHon
55
Simple Corner Example
What happens
at various
corners?
350mV
Vthp
150mV
150mV
350mV
Vthn
56
Penn ESE370 Fall2013 -- DeHon
Process Corners
• Many effects independent
• Many parameters
• Try to identify the {worst,best} set of
parameters
– E.g. Lump together things that make slow
• Vthn, Vthp, temperature, Voltage
• Try to reduce number of unique corners
– Slow corner of design space
• Use corners to bracket behavior
57
Penn ESE370 Fall2013 -- DeHon
Range of Behavior
Probability Distribution
• Still get range of performances
• Any way to exploit the fact some are faster?
Delay
58
Penn ESE370 Fall2013 -- DeHon
Probability Distribution
Speed Binning
Sell
Premium
Sell
Sell
nominal cheap
Discard
Delay
59
Penn ESE370 Fall2013 -- DeHon
Idea
• Parameters Approximate
• Differ
– Chip-to-chip, transistor-to-transistor, over
time
• Robust design accommodates
– Tolerance and Margins
– Doesn’t depend on precise behavior
Penn ESE370 Fall2013 -- DeHon
60
Midterm 1
• Contents should not be a surprise
– Identify CMOS/non-CMOS
– Identify CMOS function
– Any logic function  CMOS gate
– Noise Margins
– Circuit quasistatic configuration and
switching delay
61
Penn ESE370 Fall2013 -- DeHon
Admin
• Midterm Monday
– 7—9pm in Towne 309
• Previous midterm
– Solutions linked to 2010--2012 syllabus
• But only one midterm in 2010 so parts more
advanced than where we are now
• Review on Sunday
– 5:30pm
62
Penn ESE370 Fall2013 -- DeHon