Download PLD Applications

Reconfigurable Systems Emerge
Nick Tredennick, Editor
Gilder Technology Report
[email protected]
23 May 2017
Nick Tredennick
Overview
• Major trends affecting the microprocessor
market
– Value PC
– Value transistor
– Emerging economies
• Microprocessors
– Computer microprocessors
– Embedded microprocessors
– Configurable microprocessors
– PLD microprocessors
23 May 2017
Nick Tredennick
Moore’s Law
23 May 2017
Nick Tredennick
Top View: Field-Effect Transistor
23 May 2017
Nick Tredennick
The Microprocessor
• 10 years of Moore’s-law progress led to
the microprocessor
• The second generic component
• Raised engineers’ productivity
• Problem-solving became programming
• Grew to billions of units/year
• Stalled progress in design methods for
thirty years
23 May 2017
Nick Tredennick
The Personal Computer
• 10 years of microprocessor progress led
to the PC
• Dominated the industry for 20 years
• Supply of performance grows with
Moore’s law
• Demand grows more slowly
• Diverging growth in supply and demand
leads to the value PC
23 May 2017
Nick Tredennick
The PC Is Good Enough
23 May 2017
Nick Tredennick
The Path To The Value Transistor
23 May 2017
Nick Tredennick
Transistors Are Good Enough
23 May 2017
Nick Tredennick
Foundries: Adoption Rate By Process
Modeled after: TSMC http://www.tsmc.com/english/technology/t0203.htm
23 May 2017
Nick Tredennick
Semiconductors: Industry In Transition
• Causes
– The transistor is good enough
– The PC is good enough
• Effects
– Shift from tethered to mobile systems
• Changes design emphasis
– From: cost-performance
– To: cost-performance-per-watt
– Non-volatile memory will emerge
– Wafer stacking will emerge
– MEMS will emerge
23 May 2017
Nick Tredennick
Design Alternatives
What
• Microprocessors
• ASICs
• FPGAs
Value
$40B
$30B
$3B
Who
Programmers
Logic designers
Logic designers
FPGAs and microprocessors are
usurping a declining ASIC market.
Microprocessors (and their derivatives)
will win.
23 May 2017
Nick Tredennick
Programmers And Logic Designers
• Programmers optimize software
The Users Manual
is the
(problematic)
bridge
–
–
–
–
Languages
OS
Compilers
Applications
Programmers
• Logic designers optimize hardware
– Microprocessors
– Memory
Logic designers
23 May 2017
Nick Tredennick
Microprocessors
• Microprocessor advantages
– Flexibility
– High-volume production
– Usable by programmers
• Microprocessor limitations
– Too slow
– Too much power
23 May 2017
Nick Tredennick
Microprocessors Are Unsuitable
23 May 2017
Nick Tredennick
Application-Specific Integrated Circuits
• ASIC advantages
– Best performance
– Smallest chip
– The benchmark for function efficiency
• ASIC limitations
– Inflexible
– Expensive to design
– High fixed costs require large production
runs
– Requires logic design
23 May 2017
Nick Tredennick
Programmable Logic Devices
• PLD advantages
– Flexibility
– High-volume production
• PLD limitations
– Chips too expensive
– Too slow
– Requires logic design
23 May 2017
Nick Tredennick
ASICs and PLDs (FPGAs)
• ASICs and PLDs are competing in a
$30-billion market
– This competition will not cross into the
microprocessor market because designs
require logic designers
23 May 2017
Nick Tredennick
Supply and Demand: ASICs & PLDs
23 May 2017
Nick Tredennick
Microprocessors and ASICs
Application
Microprocessor
• For the ultimate in flexibility,
programmers map the application
onto a general-purpose
microprocessor.
Programmers
• For the ultimate in performance, logic
designers map the application into a
custom circuit.
ASIC
Logic designers
23 May 2017
Nick Tredennick
Microprocessor Evolution
Run-time
reconfigurable
microprocessor
Microprocessor
Dynamically
reconfigurable
microprocessor
Ascenium
Stretch
Programmers
Design-time
configurable
microprocessor
ARC
MIPS
Tensilica
ASIC
FPGA
23 May 2017
Logic designers
Nick Tredennick
Application
Design-Time Configurable Microprocessor
Profile the
application
Most of the
application runs as
execution of
general-purpose
instructions
Programmers
Create custom
hardware and
instructions to
accelerate critical
application sections
Design-time
configurable
microprocessor
Logic designers
23 May 2017
Nick Tredennick
Design-Time Configurable Microprocessor
•
•
•
•
Profile the application
Create custom instructions for critical code sections
Build specialized execution units
Can be 10 to100 times faster than a general-purpose
microprocessor on the target algorithm
• Examples: ARC and Tensilica
• Customized microprocessor limitations
– Requires logic designers
– Creates an application-specific, limited-function
microprocessor
– Accelerates only critical sections
23 May 2017
Nick Tredennick
Run-Time Reconfigurable Microprocessor
Most of the
application runs as
execution of
general-purpose
instructions
Application
Profile the
application
Create custom
instructions in an
FPGA fabric to
accelerate critical
application sections
Run-time
reconfigurable
microprocessor
Programmers
Logic designers
23 May 2017
Nick Tredennick
Run-Time Reconfigurable Microprocessor
• Build a general-purpose microprocessor with
integrated FPGA fabric
• Profile the application
• Create custom instructions for critical code sections
• Build custom execution units in FPGA fabric
• Can be 10 to 100 times faster than a general-purpose
microprocessor on the target application
• Example: Stretch
• Run-time reconfigurable microprocessor limitations
– Accelerates only statically identifiable critical sections
– Limited to problems for which profiling works
– Profiling is difficult
23 May 2017
Nick Tredennick
Application
Dynamically Reconfigurable Microprocessor
Create custom
instructions in a
custom fabric to
accelerate the
entire application
Dynamically
reconfigurable
microprocessor
Programmers
Logic designers
23 May 2017
Nick Tredennick
Dynamically Reconfigurable Microprocessor
• Each cycle creates a new microprocessor
implementation
– Each cycle creates a custom circuit (Ascenium instruction)
representing hundreds to thousands of conventional
instructions
• Programmed using ANSI-standard programming
languages (e.g., C/C++)
• Tens to 100s of times faster than a general-purpose
microprocessor
• Dynamically reconfigurable microprocessor
limitations
– There are none on the market today
• Until Ascenium, no one has figured out how to “program” a
dynamically reconfigurable circuit
– VCs don’t understand it
23 May 2017
Nick Tredennick
Microprocessors
•
•
•
•
•
•
•
•
•
•
23 May 2017
x86
ARC
ARM
MicroBlaze
MIPS
Nios
PowerPC
SPARC
Tensilica
Old stuff
AMD, Intel, Transmeta, Via
ARC
ARM
Xilinx
MIPS
Altera
IBM, Freescale
Sun
Stretch, Tensilica
Everyone
Nick Tredennick
Microprocessor Applications
•
•
•
•
Supercomputers
Workstations and servers
PCs
Embedded systems
–
–
–
–
–
–
23 May 2017
Automobiles
Cameras
Cell phones
Game players
MP3 players
Set-top boxes
Nick Tredennick
Computer Markets
Millions of Units
180
160
140
120
100
80
60
40
20
0
Windows
23 May 2017
Servers
Nick Tredennick
SuperCs
Microprocessor Markets
Millions of Units
8000
7000
6000
5000
4000
3000
2000
1000
0
Embedded
23 May 2017
PCs
Nick Tredennick
Servers
Computer Microprocessors
• x86
– AMD
– Intel
– Transmeta
– Via
• Proprietary
– IBM
– Freescale
– Sun
23 May 2017
Nick Tredennick
Embedded Microprocessors
•
•
•
•
x86
ARM
PowerPC
Old stuff
AMD, Transmeta, Via
ARM
IBM, Freescale
Everyone
– Triscend (Xilinx)
23 May 2017
Nick Tredennick
Configurable Microprocessors
•
•
•
•
•
23 May 2017
ARC
Ascenium
MIPS
Nios
Tensilica
ARC
Ascenium
MIPS
Altera
Stretch, Tensilica
Nick Tredennick
PLD Microprocessors
• Altera
– Nios (soft)
• Xilinx
– MicroBlaze (soft)
– PicoBlaze (soft)
– PowerPC (hard)
23 May 2017
Nick Tredennick
Microprocessor-like
• DSPs
• Network processors
• Specialty processors
23 May 2017
Nick Tredennick
ASICs & Microprocessors
23 May 2017
Nick Tredennick
ASICs & Microprocessors
23 May 2017
Nick Tredennick
ASICs & Microprocessors
23 May 2017
Nick Tredennick
Semiconductor Trends
• Value PCs outsell leading-edge PCs
– Mobile applications emerge
– Design emphasis shifts from cost
performance to cost-performance-per-watt
• Value transistors outsell leading-edge
transistors
– Transistor performance overshoots many
applications
– Increasing demand in emerging economies
– Foundry strength grows
23 May 2017
Nick Tredennick
Consequences
• Rise of mobile applications
– New non-volatile memories
• Rise of foundries
– Rise of soft (IP) cores
– Horizontal fragmentation of integrated device
manufacturers
• Rise of non-volatile FPGAs
• Rise of reconfigurable systems
• Growing market for embedded microprocessors
– Tethered: traditional role
– Mobile: supervisory role
23 May 2017
Nick Tredennick
Industries in Transition
•
•
•
•
•
•
23 May 2017
Automotive
Analog
Mechanical
Isolated
►
►
►
Digital
Electrical
Connected
Analog
Copper
►
►
Digital
Optical, Wireless
Analog
Wet labs
►
►
Digital
Bioinformatics
Analog
Isolated
►
►
Digital
Connected
Analog
Isolated
Tethered
►
►
►
Digital
Connected
Untethered
Desktop
►
Embedded
Telecom
Biomedical
Film/video
Consumer
Computers
Nick Tredennick