Great job on projects Final next week Friday 5-8m in 166...
Transcript of Great job on projects Final next week Friday 5-8m in 166...
![Page 1: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/1.jpg)
EE141
1
EE141 EECS141 1 Lecture #30
EE141 EECS141 2 Lecture #30
Great job on projects
Final next week Friday 5-8m in 166 Burrows Covers all material
Homework 10 (arithmetic) and it solutions posted this weekend
Intended for practice only
Not graded
![Page 2: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/2.jpg)
EE141
2
EE141 EECS141 3 Lecture #30
EE141 EECS141 4 Lecture #30
![Page 3: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/3.jpg)
EE141
3
EE141 EECS141 5 Lecture #30
EE141 EECS141 6 Lecture #30
![Page 4: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/4.jpg)
EE141
4
EE141 EECS141 7 Lecture #30
EE141 EECS141 8 Lecture #30
Optimization constraints different than in
binary adder
Once again:
– Need to identify critical path
– And find ways to use parallelism to reduce it
Other possible techniques
Logarithmic versus linear (Wallace Tree Mult)
Data encoding (Booth)
Pipelining
First glimpse at system level optimization
![Page 5: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/5.jpg)
EE141
5
EE141 EECS141 9 Lecture #30
EE141 EECS141 10 Lecture #30
Technology
Node (nm)
45 32 22 16 11 8
Integration
Capacity (BT)
8 16 32 64 128 256 512 1024
Delay Scaling >0.7 ~1?
Energy Scaling ~0.5 >0.5
Transistors Planar 3G, FinFET
Variability High Extreme
ILD ~3 towards 2
RC Delay 1 1 1 1 1 1
Metal Layers 8-9 0.5 to 1 Layer per generation
THE OPTIMISTIC PERSPECTIVE
![Page 6: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/6.jpg)
EE141
6
EE141 EECS141 11 Lecture #30
Energy!
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0.11
0.12
20 30 40 50 60 70 80 90
Technology node (nm)
EO
P (
fJ)
Silicon Process Technology
Co
st
cost per
transistor
product
cost
Further scaling
is not profitable
Cost
Size
EE141 EECS141 12 Lecture #30
![Page 7: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/7.jpg)
EE141
7
Infrastructural
core
Sensory
swarm
Mobile
access
THE IT PLATFORM OF THE NEXT DECADE
TRILLIONS OF
CONNECTED DEVICES [J. Rabaey, ASPDAC’08]
It Is All About Energy …
![Page 8: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/8.jpg)
EE141
8
Data and Compute Centers “The IT workhorses”
Doing Nothing
Well!
Major Opportunity is in Power Management
Requires Top-Down System Level Solution
[Barroso, Holzle, 20
Mobiles “The home of the user interface” Most “tricks” already in use! (multi-core,
heterogeneity, accelerators, SoC, …)
Mobile μProc Anno 2015 [Courtesy A. Peleg, Intel]
Opportunity: system and
application considerations
Always-connected
Perceptual processing
![Page 9: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/9.jpg)
EE141
9
The Sensory Swarm “Adding senses to the Internet”
Philips Sand module
UCB mm3 radio
UCB PicoCube
[Ref: Ambient Intelligence, W. Weber Ed., 2005]
IMEC e-Cube
Telos Mote
The driver for
Ultra-Low Energy
design for past decade
Yet … True Immersion Still Out of Reach Microscopic Wireless
Artificial Skin
“Microscopic” Health Monitoring
Interactive Surfaces
Another leap in size, cost and energy reduction
Smart Objects
![Page 10: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/10.jpg)
EE141
10
Example: Microscopic Wireless to Power Brain-Machine Interfaces (BMI) The Age of Neuroscience BMI – The Instrumentation of Neuroscience
• Learning about operation of the brain
• Enabling advanced prosthetics
• Enabling innovative human-machine
interfaces
mm3 nodes
remotely powere
uWs to 1 mW power budget
The Dream: Observing Living Cells
Combines sensing,
processing, communication
and energy harvesting and storage in volumes far less
than 1 mm3
10-100
μm
[Courtesy, Hillenius et al, 07]
![Page 11: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/11.jpg)
EE141
11
The Holy Grail: Reducing the Energy/Operation
CMOS digital
Non-CMOS
digital
Interfaces and periphery
Long term musings
Energy Limits in Digital
Claude Shannon John Von Neumann
More than 4 orders of magnitude
below current practice (65 nm at 1V)
Shannon-Von Neumann-
Landauer Bound:
Minimum energy/
operation = kTln(2)
= 4.10-21J/bit at room
temperature
![Page 12: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/12.jpg)
EE141
12
Technology Scaling Not the Solution
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0.11
0.12
20 30 40 50 60 70 80 90
Technology node (nm)
EO
P (
fJ)
[Based on actual and predictive models]
Sub-Threshold Operation Leads to Minimum Energy/Operation
Op
timal (V
dd , V
th )
Threshold Voltage (Vth)
Supp
ly V
olta
ge
(VD
D)
Energy-Aware FFT Processor
[Chang, Chandrakasan, 2004]
Energy
self-contained processors
Subliminal μprocessor for
retinal implants
3 pJ/inst @ 350 mV
[Blaauw, VLSI’07]
![Page 13: Great job on projects Final next week Friday 5-8m in 166 ...bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_s09/Lectures/Lectu… · 8 16 32 64 128 256 512 1024 Delay Scaling >0.7](https://reader033.fdocuments.us/reader033/viewer/2022060606/605b482f4ee2a36c425d081f/html5/thumbnails/13.jpg)
EE141
13
How About Mechanical Computing?
NEMS Relay
[Courtesy: TJ King, E. Alon, UCB]
ON: VGS > VPI – Low on resista
OFF: VGS < VPI – Zero Leakag
NEMS Relays Versus CMOS
32-bit adder CMOS Relay
Supply
Voltage
0.5 V 0.32 - 0.9 V
Load Cap
per Output
25 fF 25 fF
Total Gate
Cap
4.0 pF 125 fF
Area 600 μm2 480 μm2
[CMOS Adder: D. Patil, ARITH'07]
9x
10x
Vdd: 1V 0.5V
Energy/op vs. Delay/op across Vdd
Enables the parallelism concept anew!