DIC - Lec1 - Overview Design Methodology
DIC - Lec1 - Overview Design Methodology
2
Grading of this Course
60% Labs
Tools: Verilog, HSPICE
Most exercises are related to circuit-level implementation
20% Final Team Project
20% Final Exam
On-line choice questions
3
Undergraduate Course v.s. Graduate Course
Undergraduate courses (VLSI)
Basic transistor and circuit models
Basic circuit design styles
First experiences with design – creating a solution with given
numbers of specifications
Graduate courses
Transistor models of varying accuracy
Design under constraints: power-constrained, flexible,
robust,…
Learning the more advanced techniques
Study the challenges facing design in the near future
Creating new solutions to challenging design problems
4
Digital Processing or Digital Computing
Analog real world v.s. digital computing system
Discrete digital data (binary digits: 0 & 1)
5
Signal Characteristics
Analog signals
DC + Small signals (current-based or voltage-based)
6
Binary Digital Signal
An information variable represented by physical quantity.
Binary values are represented abstractly by:
Digits 0 and 1
Words (symbols) False (F) and True (T)
Words (symbols) Low (L) and High (H)
And words on and off
Binary values are represented by ranges of values of physical
quantities.
7
Digital or Analog Signal?
9
Introduction to Signal Aberration
Power noises (power integrity)
Parasitic R,L,C
Coupling effect
Non-continuous physical links
10
2-level Circuits
11
2-level Circuits
12
Transistors as Switches
We can view MOS transistors as electrically
controlled switches
Voltage at gate controls path from source to
drain
g=0 g=1
d d d
nMOS g OFF
ON
s s s
d d d
pMOS g OFF
ON
s s s
13
What is Transistor?
VGS ≥ V T |VGS|
Ron
S D
14
CMOS Inverter
A Y VDD
0 1
1 0 OFF
ON
0
1
A Y
ON
OFF
A Y
GND
15
CMOS NAND Gate
A B Y
ON
OFF
OFF
ON OFF
ON
0 0 1
0 1 1
1
0
Y
1 0 1
A
ON
OFF
1 1 0 0
1
1
0
B
OFF
ON
ON
OFF
16
CMOS NOR Gate
A B Y
0 0 1 A
0 1 0
1 0 0 B
1 1 0 Y
17
Basic Logic for Digital System in CMOS ICs
Combination logic
Control switch (MUX)
Computation (XOR)
Basic logic – INV, NAND, NOR
Sequential circuits & Finite State Machine (FSM)
Storage element (Latch, Flip-Flop)
Data storage
Register file, SRAM, eDRAM, eFlash…
I/O (input/output)
ESD protection, driving ability, level-sensing
18
The Regenerative Property for SRAM
Vi1 Vo1 Vi2 Vo2 C
cascaded inverters A B
A
If the gain in the transient
region is larger than 1,
C
only A and B are stable
operation points. C is a
B metastable operation
Vi1 = Vo2 point.
19
How to put together millions of transistors?
Well chosen design methodologies
Well chosen architectures
Extensive use of power full CAD tools
Strict design management
Well chosen testing methodologies
Design re-use
One can NOT use same design methodologies and
architectures when complexity increases orders of
magnitude
20
General Principles
Technology changes fast => important to
understand general principles
optimization, tradeoffs
work as part of a group
leverage existing work: programs ,building blocks
Concepts remain the same:
Example: relays -> tubes -> bipolar transistors
->
MOS transistors -> FinFET -> GAAFET (MBCFET)
Future Trend for advanced
technology nodes or 3DIC
DTCO: Design-technology co-optimization
21
MOS Transistor: 3D Perspective
22
FinFET
23
Complementary Charastertcis
24
GAAFET, MBCFET, RibbonFET
Transistors have evolved to provide better PPA
for designers
GAA is the ultimate transistor with 4-sided
channel
GAA flexible Weff provides more freedom to
optimize PPA
25
Design Abstraction Levels
SYSTEM
Top-Down
Design
MODULE
+
GATE
CIRCUIT
Bottom-Up
Library G
DEVICE
S D
n+ n+
26
Typical IC Design Flow (Methodology)
27
Foundry (Process) Design Kit (FDK, PDK)
28
Full-Custom Design
Design some or all of the circuits, logic cells,
layout specifically
Required cells/IPs are not available
Existing cell libraries can not meet the requirements
Area, speed or power consumption
Technology migration (analog or mixed-mode
design)
Demand long design time
29
Cell-based (IP-Based) Design
Use pre-designed logic cells (known as standard
cells) and micro cells or IPs
Each standard cell can be optimized individually
All mask layers are customized
Custom blocks can be embedded
30
System Design Paradigm
Data computation (functional IP)
Data communication (platform)
Data storage (memory)
31
SoC Hardware Design Methodology
32
Design Metrics
Evaluate performance of a digital circuit
and System
Cost
Speed (delay, operating frequency)
Power dissipation
Energy to perform a function
Robustness
Reliability
Scalability
33
Design Challenges for Future IC Design
Large variations
Increasing leakage current
Increasing RC of Interconnects
Power integrity (large power noises)
Voltage with
dynamic IR drop
Minimal
Tolerance Level
[IEDM’19]
Large chip
Small chip
Gate delay