Upload
phamquynh
View
224
Download
0
Embed Size (px)
Citation preview
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 1
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
Directions of Programming Research:
Seeking a Needle in the Haystack?
Reiner Hartenstein
IEEE fellow
Keynote, 1st Brazilian-German Workshop on Micro and Nano
Electronics (BGME’2010), Oct 6-8, 2010, Porto Alegre, RS, Brazil
1
Brasil-Alemanha 2010 / 2011: Ano da Ciência, Tecnologia e Inovação,
Karlsruhe Institute of Technology
member,
German-Brazilian Year of Science - Technology and Innovation 2010/2011,
Deutsch-Brasilianisches Jahr der Wissenschaft, Technologie und Innovation 2010/2011
http://hartenstein.de/BGME-10.pdf © 2008, [email protected] http://hartenstein.de 2010, 2010,
Abstract (Preface)
2
But the Energy Consumption of Computing worldwide will become unaffordable.
We need to reinvent Computing.
Reconfigurable Computing (RC) could be a Silver Bullet to solve such Problems
And the Performance Progress is stalled by the Parallelism Wall and the Power Wall
However all this is massively handicapped by Programmer Productivity Problems etc.
We need more Computing Services
© 2008, [email protected] http://hartenstein.de 2010, 2010, 3
The Twin Wall Crisis:
Power & Performance
Worldwide two drastically disruptive Developments:
µP Industry changed Strategy over to „manycore“
(away from faster Clock Speed)
A Variety of Power Consumption Problems
The Programming Wall
The Power Wall
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Outline
• The coming Shortage of Energy
• Why Computing is important
• Energy Consumption of Computing
• The Programmability Crisis
• Rescue by Reconfigurable Computing ?
• The Reconfigurability Paradox
• We need to Reinvent Computing
• Reinventing Programmer Education
• Conclusions
4
© 2008, [email protected] http://hartenstein.de 2010, 2010,
No more cheap oil
5
Currently: >80 $
Tendency: growing
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Oil crises: weekend
driving ban (Germany)
6
1973 1979/1980
(depencence on near east oil countries)
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 2
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Beyond Peak Oil
7
J. S. Gabrielli de Azevedo: Petrobras
e o Novo Marco Regulatório;
São Paulo, December 1, 2009
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Cheap Oil Era reached its End
Rapidly growing energy prices (IEA: factor of 3) predicted.
50% Reserves are under Water. Off-shore Projects re-calculated.
IEA: “need six more Saudi Arabias for Demand predicted for 2030“
80% of Crude Oil is coming from Decline Fields.
Higher Standards of living: China, India, Brazil, Mexico, newly Industr. Countries.
R.H.: Growing Power Consumption of Computers: 10 more Saudi Arabias!
IEA estimates: Demand will at least double til the Year 2030
China passes the U.S. in Energy use [IEA]
8
© 2008, [email protected] http://hartenstein.de 2010,
Beyond Oil: Literature
9
US: ~3 $
… post petroleum …
… hundreds of books
© 2008, [email protected] http://hartenstein.de 2010,
Beyond oil:
Literature (2)
10
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Outline
• The coming Shortage of Energy
• Why Computing is important
• Energy Consumption of Computing
• The Programmability Crisis
• Rescue by Reconfigurable Computing ?
• The Reconfigurability Paradox
• We need to Reinvent Computing
• Reinventing Programmer Education
• Conclusions
11 © 2008, [email protected] http://hartenstein.de 2010, 12
Banking without Computers
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 3
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Business Information Systems
without Computers
13
Lufthansa Reservation
anno 1960
http://wiki.answers.com/Q/Why_are_computers_important_in_the_world
[Courtesy Ernst
Denert]
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Computing is important
Survival and Success of the global Economy
14
We have to convince all Funding Sources
© 2008, [email protected] http://hartenstein.de 2010,
Computers everywhere
15 © 2008, [email protected] http://hartenstein.de 2010, 16
... Ecosystem: just one example
16
© 2008, [email protected] http://hartenstein.de 2010, 2010,
... Supercomputers ...
17 © 2008, [email protected] http://hartenstein.de 2010, 2010,
Data Centers ...
18
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 4
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010, 19
more to come ...
… more to come and more needed
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Innovation-driven computing
[Andy Hopper]
• Simulation and Modelling are important Tools which will help predict Global Warming and its Effects.
20
• Computing will play a Key Part in optimizing use of Resources in the physical World.
• The Amount of Infrastructure making up the Digital World is continuing to grow rapidly and starting to consume significant Energy Resources.
• To help generate Momentum and achieve these Goals, it is important that a coordinated set of challenging international Projects are investigated.
• We are experiencing a Shift to the digital World in our daily Lives as witnessed by the wide scale Adoption of the World Wide Web.
Green IT:
• Smart energy Meters: Housing, Buildings, Facilities
• Carpooling and public Transport by Info Web Sites
• Road Traffic and Transport Logistics Optimization
• Reduce Travelling by Telecommuting.
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Some Grand Challenge
Examples for CPS*
21
[Ed. Lee]
• Blackout-free electricity generation and distribution,
• Extreme-yield agriculture,
• Safe, rapid evacuation in response to natural or man-made disasters,
• Perpetual life assistants for busy, senior/disabled people,
• Location-independent access to world-class medicine,
• Near-zero automotive traffic fatalities, minimal injuries, and significantly reduced traffic congestion and delays,
• Reduce testing and integration time and costs of complex CPS systems (e.g. avionics) by 1 to 2 orders of magnitude,
• Energy-aware buildings and cities,
• Physical critical infrastructure that calls for preventive maintenance,
• Self-correcting cyber-physical systems for “one-off” applications,
• Disaster Response: Large-Scale Emergency Evacuation,
• Assistive Devices.
*) Cyber-Physical Systems http://varma.ece.cmu.edu/Summit/
© 2008, [email protected] http://hartenstein.de 2010, 2010,
The World Economic Forum’s
"Global Redesign Initiative”
Organizations like UN, UNESCO; GATT, G8, G20 are increasingly inept at fixing what ails the world:
22
• economic growth • climate protection • poverty eradication • conflict avoidance • human security • global vaccine protocol • global risk management • promotion of shared values • intelligent water management • smart energy production/distribution …
“Existing global institutions require extensive rewiring to confront contemporary challenges."
Wikinomics approach for agile world-wide mass collaboration without bureaucracy.
for citizen juries, polling, digital brainstorms, policy
wikis, town hall meetings …
New paradigm to involve world citizens by global IT networks
with graphic user interfaces
Doha,
Qatar,
30-31 May
2010
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Growing Number of Computers
Trends [Glesner's List]: • Security • Mobility • Communication • Health • Bio-medical • Support of the Elderly • Quality of Life • Leisure • Environment • Sustainable Development • Energy
23
List by Rene Penning de Fries: •Energy •Water •Food •Environment •Poverty •Terrorism •War •Disease •Education •Democracy •Population (2050: ~10 billion people)
[Courtesy Manfred Glesner]
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Outline
• The coming Shortage of Energy
• Why Computing is important
• Energy Consumption of Computing
• The Programmability Crisis
• Rescue by Reconfigurable Computing ?
• The Reconfigurability Paradox
• We need to Reinvent Computing
• Reinventing Programmer Education
• Conclusions
24
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 5
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Growth of the Internet
The trends are illustrated by :
expanding wireless internet,
growing number of users.
shipping electronic books,
more cloud computing?
and many other services.
25
2007 to 2030 factor x30 predicted
Broadband connections NA, Mex, WE by end’ 2007: 155 millions - predicted for after 2011: 228 millions.
larger e-mails,
services integr’g video and software
increasing popularity of games,
massive use of video on demand,
high-definition video and pay-TV,
services by mobile phone companies.
[G. Fettweis, E. Zimmermann: ICT Energy Consumption - Trends and Challenges; WPMC'08, Lapland, Finland, 8 –11 Sep 2008]
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Power Consumption of Computers
[Albert Zomaya 2008]
26
at Dallas
[Randy Katz: IEEE Spectrum, Febr. 2009]
Energy cost may overtake
IT equipment cost in the near future
„Google causes 2% of the worlds
electricity consumption“
(Google denied)
at Quincey
at Boardman
2009
Exascale (1018) supercomputer predicted for ~2020: 120 MW
“Largest supercomputers, including exaflops, should not be thought of as computers. They’re strategic scientific instruments. Their usage patterns and scientific impact are closer to major facilities such as CERN, ITER, or Hubble.” [Andrew Jones, Numerical Algorithms Group, ZDNet]
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Electricity Bill: a Key Issue
„The possibility of computer equipment power consumption spiraling out of control could have serious consequences for the overall affordability of computing.”
Patent for water-based data centers
Google data center cost determined by monthly power bill
[L. A. Barrosso, Google, 2009]
27
Google going to sell electricity
• Already 2005, Google’s electricity bill higher than value of its equipment.
© 2008, [email protected] http://hartenstein.de 2010, 2010, 28
How Societies Chose
to Fail or Succeed
Unaffordability of von-Neumann-centric computing could jeopardize all facets of our global economy.
Manycore: failure could jeopardize both, IT industry & most sections of the economy depending on rapid improvement of IT. [Dave Patterson]
Several recent outages of cloud computing services.
Stuxnet worm: only propaganda trick?
© 2008, [email protected] http://hartenstein.de 2010, 2010,
without Cyber Infrastructure ?
29
?
copyrighted!
? ?
homo computensis
homo Neanderthalensis?
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Outline
• The coming Shortage of Energy
• Why Computing is important
• Energy Consumption of Computing
• The Programmability Crisis
• Rescue by Reconfigurable Computing ?
• The Reconfigurability Paradox
• We need to Reinvent Computing
• Reinventing Programmer Education
• Conclusions
30
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 6
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
The Trouble with Manycore
The growing core counts are racing ahead of programming paradigms and programmer productivity
31
- a challenge to CS education
to major extent also in mass markets
going to FPGA: for programmers a paradigm shift
in supercomputing &
Chipmakers busy designing microprocessors that most programmers can’t program … [David Patterson, IEEE Spectrum, July 2010]
… doing so without any clear notion of how such devices would in general be programmed
They hope, someone will be able to figure out how to do that
© 2008, [email protected] http://hartenstein.de 2010, 2010,
The vast majority of HPC or supercomputing applications originally written for single processor with direct access to main memory.
32
Programmability Crisis
But the first petascale supercomputers employ more than 100,000 processor cores each, and distributed memory.
They hope, that dozens of applications are inherently parallel enough to be laboriously decomposed, sliced and diced, for mapping onto HPC
Large applications only modestly scalable. >50% apps don’t scale beyond 8 cores, only 6% can exploit >128 PE cores, a tiny fraction 100,000 or more cores.
we are in an all-dominant
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Multicore is not new
•ACRI •Alliant •American Supercomputer •Ametek •Applied Dynamics •Astronautics •BBN •CDC •Convex •Cray Computer •Cray Research •Culler-Harris •Culler Scientific •Cydrome •Dana/Ardent/ Stellar/Stardent
•DAPP •Denelcor •Elexsi •ETA Systems •Evans and Sutherland Computer •Floating Point Systems •Galaxy YH-1 •Goodyear Aerospace MPP •Gould NPL •Guiltech •ICL •Intel Scientific Computers •International Parallel Machines
•Kendall Square Research •Key Computer Laboratories
Dead (Super)Computer Society [Gordon Bell, keynote, ISCA 2000]
•MasPar •Meiko •Multiflow •Myrias •Numerix •Prisma •Tera •Thinking Machines •Saxpy •Scientific Computer •Systems (SCS) •Soviet Supercomputers •Supertek •Supercomputer Systems •Suprenum •Vitesse Electronics
the single core sequential mind set was the winner
only 2 or 3 successes
most in 1985-1995
- mainly research
33
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Can we get it right this time?
The “parallel programming problem”: has been addressed for at least 25 years.
34
Only a small number of specialized developers can write parallel code.
[T. Mattson, M. Wrinn: Parallel Programming:
Can we PLEASE get it right this time? DAC 2008, Anaheim, CA, June 8-13, 2008],
A massive worldwide effort is required, taking many years, creating masses of jobs We need to reinvent
programmer education
We need to reinvent computing
„The proud era
of von Neumann architecture passes into history.“
„Foundational change will disrupt traditional habits throughout the discipline ....“
[Michael Wrinn, (keynote, SIGCSE2010): Suddenly, All Computing is Parallel:
Seizing Opportunity Amid the Clamor] http://www.sigcse.org/sigcse2010/attendees/keynotes.php
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Amid the Clamor
35
How to
bring parallel
computing into
mainstream of
undergraduate
education ?
[Michael Wrinn]
current discussions: despairingly seeking a needle in a haystack.
© 2008, [email protected] http://hartenstein.de 2010, 2010,
We need a new Textbook
36
having an impact like Mead & Conway
"The book that changed everything“; Electronic Design News, Feb. 11, 2009
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 7
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Outline
• The coming Shortage of Energy
• Why Computing is important
• Energy Consumption of Computing
• The Programmability Crisis
• Rescue by Reconfigurable Computing ?
• The Reconfigurability Paradox
• We need to Reinvent Computing
• Reinventing Programmer Education
• Conclusions 37 © 2008, [email protected] http://hartenstein.de 2010, 2010,
The Tail wagging the Dog
38
„Central“: it controls
(almost) everything
However,
it needs
accelerators
accelerators CPU
CPU „Central Processing Unit“
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Twin paradigm systems
39
CPU „Central Processing Unit“
„Central“: it controls
(almost) everything
However,
it needs
accelerators
CPU
hardwired
accelerators
reconfigurable
accelerators second paradigm
accelerators
ASIC: 3%
FPGA: 97%
[Dataquest, 2009]
design start ratio
© 2008, [email protected] http://hartenstein.de 2010, 2010, 40
A Clean Terminology, please
program source result
Software instruction streams
Flowware data streams
Configware datapath structures configured
© 2008, [email protected] http://hartenstein.de 2010, 2010, 41
FFT
100
Reed-Solomon Decoding 2400
Viterbi Decoding 400
1000
MAC
DSP and wireless
molecular dynamics simulation
88
BLAST 52
protein identification
40
Smith-Waterman pattern matching
288
Bioinformatics
GRAPE
20 Astrophysics
SPIHT wavelet-based image compression 457
real-time face detection
6000
video-rate stereo vision
900
pattern recognition 730
Image processing, Pattern matching, Multimedia
3000 CT imaging crypto
1000
28500
DES breaking
1
1000
1,000,000
Spe
edup
-Fac
tor
Speed-up
factors
obtained
by Software
to Configware
migration
8723 DNA seq.
100
10
10,000
100,000
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Energy saving factors: ~10% of speedup
42
FFT
100
Reed-Solomon Decoding 2400
Viterbi Decoding 400
1000
MAC
DSP and wireless
molecular dynamics simulation
88
BLAST 52
protein identification
40
Smith-Waterman pattern matching 288
Bioinformatics
GRAPE
20 Astrophysics
crypto 1000
28500 DES breaking
1
1000
1,000,000
Spe
edup
-Fac
tor
Power save
factors
obtained
(FPGAs) SPIHT wavelet-based
image compression 457
real-time face detection
6000
video-rate stereo vision
900
pattern recognition 730
Image processing, Pattern matching, Multimedia
3000 CT imaging
8723 DNA seq.
100,000
10,000
100
10
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 8
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
FPGA vs.x86 Clock Frequency
43
1995 2000 2005100
1
10GHz
GHz
MHz
Pentium IV
Pentium III
Celeron
Celeron to Pentium IV: x20
growing clock speed
growing power consumption
(migration papers: power save not reported before 2005)
Pentium I (1989) to Pentium IV: x60
1995 – 2005: speed-ups obsolete?
-> not obsolete !!
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Hitting 28nm, and beyond
Both de facto FPGA giants (Xilinx and Altera) are hitting 28nm at end of 2010.
44
2009: Intel ships 32nm,
2010: foundries to ship 28nm
Intel will ship 22 nm in 2011,
16 nm in 2013
Also FPGAs benefit from Technology progress.
New Xilinx and Altera products push the power limits also for FPGA use in battery-powered small portables
(IGLOO from Altera, from Xilinx: 28 nm low-power process at TSMC and Samsung Foundry)
© 2008, [email protected] http://hartenstein.de 2010, 2010, 45
[Tarek El-Ghazawi et al.: IEEE COMPUTER, Febr. 2008]
Application . Speed-up factor
Savings Power Cost Size
DNA and Protein sequencing 8723 779 22 253
DES breaking 28514 3439 96 1116
much less equipment
needed
massively saving energy
RC*: Demonstrating the intensive Impact
SGI Altix 4700 with RC 100 RASC compared to Beowulf cluster
Tarek El-Ghazawi
*) RC = Reconfigurable Computing © 2008, [email protected] http://hartenstein.de 2010,
Drastically less Equipment needed
For instance: a hangar
full of racks replaced by a single rack
without air
conditioning
46
or ½ rack
© 2008, [email protected] http://hartenstein.de 2010, 2010, 47
END
SGI® RASC™ Module (Version1) Xilinx Virtex II-6000 FPGA 16MB QDR SRAM Rack-mountable Dual NUMAlink™ 4 ports Seamless direct attach to server's shared memory fabric Datasheet (PDF 145K)
SGI® RASC™ RC100 Blade Dual Virtex 4 LX200 FPGAs 80MB QDR SRAM or 20GB DDR2 SDRAM Blade or rack-mountable form factor Dual NUMAlink™ 4 ports Seamless direct attach to server's shared memory fabric Datasheet (PDF 137K)
Hetero HPC
SGI® RASC™
© 2008, [email protected] http://hartenstein.de 2010, 2010,
The silver bullet
Reconfigurable Computing is really the silver bullet for massively saving energy
48
scene „Green Computing“ Reinvent Computing
predicted energy saving
factor of about 3 orders of magnitude
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 9
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Outline
• The coming Shortage of Energy
• Why Computing is important
• Energy Consumption of Computing
• The Programmability Crisis
• Rescue by Reconfigurable Computing ?
• The Reconfigurability Paradox
• We need to Reinvent Computing
• Reinventing Programmer Education
• Conclusions
49 © 2008, [email protected] http://hartenstein.de 2010, 2010,
FPGA: much worse Technology?
Lower Clock Frequency (x3 – x5)
50
huge Wiring Overhead (for routing)
Reconfigurability Overhead: e. g. only about 4 % of Transistors serve the Application
& Routing Congestion …
Speed-up by Orders of Magnitude with an Orders of Magnitude worse Technology?
The Reconfigurability Paradox
© 2008, [email protected] http://hartenstein.de 2010, 51
the tremendous inefficiency of
computers causes immense
electricity consumption
the tremendous inefficiency of
computers causes immense
electricity consumption
51
because
of The von Neumann
Syndrome
© 2008, [email protected] http://hartenstein.de 2010, 2010,
All but ALU is overhead: x20 efficiency
52
(data cashe)
x20
inefficiency:
just one
of several
overhead
layers
[R. Hameed et al.: Understanding Sources of Inefficiency in General-Purpose Chips; 37th ISCA, June 19-23, 2010, St. Malo, France]
© 2008, [email protected] http://hartenstein.de 2010,
Massive Overhead Phenomena
proportionate to the number of processors
overproportionate to the number of processors
53
overhead von Neumann
machine
instruction fetch instruction stream
state address computation instruction stream
data address computation instruction stream
data meet PU + other overh. instruction stream
i / o to / from off-chip RAM instruction stream
Inter PU communication instruction stream
message passing overhead instruction stream
transactional memory overh. instruction stream
multithreading overhead etc. instruction stream
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Critique of the von Neumann Model
Brad Cox 1990:
Planning the Software
Industrial Revolution
Dijkstra 1968: The Goto considered harmful
R. Hartenstein, G. Koch 1975: The universal Bus considered harmful
Backus 1978: Can programming be liberated from the von Neumann style
Arvind et al., 1983: A critique of Multiprocessing the von Neumann Style L. Savain 2006: Why Software is bad
Critique of von Neumann is not new:
Peter G. Neumann 1985-2003: 216x “Inside Risks“ 18 years inside back cover of Comm_ACM
Peter G. Neumann
54
overhead piles up to code sizes
of astronomic dimensions
“von Neumann
Syndrome”: C.V. Ramamoorthy; UC Berkeley
Nathan’s Law: Software is a gas.
It expands to fill all its containers ...
Nathan Myhrvold
Wirth‘s Law
“software is slowing faster
than hardware is accelerating“
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 10
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010, 55
Software is too easy to change: Overhead, Featuritis, and: Bugs
Featuritis & Overhead
Configware is less easy to change: a highly important advantage ?
Robert N. Charette: Why Software Fails; IEEE Spectrum, Sep 2005
© 2008, [email protected] http://hartenstein.de 2010,
56
von Neumann overhead vs.
Reconfigurable Computing
overhead von Neumann
machine datastream machine
instruction fetch instruction stream none*
state address computation instruction stream none*
data address computation instruction stream none*
data meet PU + other overh. instruction stream none*
i / o to / from off-chip RAM instruction stream none*
Inter PU communication instruction stream none*
message passing overhead instruction stream none*
transactional memory overh. instruction stream none*
multithreading overhead etc. instruction stream none*
56
© 2008, [email protected] http://hartenstein.de 2010, 2010,
The History of Computing
(1)
57
The 1st electrical computer, ready
prototyped for mass production ?
Guess: which year, which company ?
© 2008, [email protected] http://hartenstein.de 2010, 58
The History of Computing
(2) Prototype 1884: Herman Hollerith
datastream-based
the first reconfigurable computer
DPU
The first Xilinx FPGA came 100 years later
size: like about 3 refrigerators
1989 used for US population census
© 2008, [email protected] http://hartenstein.de 2010, 59
Early LUT
60 years later: RAM available – e. g. ferrite core
non-volatile configuration memory
field-programmable:
manually
swapping plug boards
(motivation for von Neumann machine paradigm)
© 2008, [email protected] http://hartenstein.de 2010,
60 years later
60
much larger than 3 refrigerators
just for a few ballistic tables:
the „von Neumann“ paradigm
von Neumann syndrome
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 11
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Outline
• The coming Shortage of Energy
• Why Computing is important
• Energy Consumption of Computing
• The Programmability Crisis
• Rescue by Reconfigurable Computing ?
• The Reconfigurability Paradox
• We need to Reinvent Computing
• Reinventing Programmer Education
• Conclusions
61 © 2008, [email protected] http://hartenstein.de 2010,
62
term controlled by execution
triggered by paradigm
CPU program counter
(at ALU)
instruction fetch
instruction stream
DPU**
rDPU**
data counter(s) (at memory)
data arrival* data-stream-based
*) “transport-triggered” **) does not have a program counter
- no instruction fetch
Single Paradigm (from the
Mainframe Age) is obsolete
© 2008, [email protected] http://hartenstein.de 2010,
63
term controlled by execution
triggered by paradigm
CPU program counter
(at ALU)
instruction fetch
instruction stream
DPU**
rDPU**
data counter(s) (at memory)
data arrival* data-stream
+ New Machine Model for FPGAs
*) “transport-triggered” **) does not have a program counter
- no instruction fetch
twin paradigm
twin paradigm
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Acceleration Mechanisms
parallelism by multi bank memory architecture auxiliary hardware for address calculation address calculation before run time
avoiding multiple accesses to the same data. avoiding memory cycles for address computation optimization by storage scheme transformations optimization by memory architecture transformations
Accelerate tasks by streaming
Can achieve 100x in improved use of memory.
MISD structured computation: streaming computations across a long array before storing results in memory.
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Programming Model: Flowware
Adder
Speaker
FMDemod
LPF1
Split
Gather
LPF2 LPF3
HPF1 HPF2 HPF3
Source:
MIT
StreamIT
• Pros for streaming – Streamlined, low-overhead
communication – (More) deterministic behaviour – Good match for many simple media
rich applications
[Pierre Paulin]
We‘ve to find out, which applications types and programming models Students should exercise for the flowware approach
• Cons – control-dominated applications – shunt yard problem
© 2008, [email protected] http://hartenstein.de 2010, 2010,
The Machine Model Dichotomy
(1) von Neumann versus data stream machine
66
PE
Program Engineering
*) do not confuse with „dataflow“!
Flowware
Engineering
FE
auto-sequencing Memory
asM
SE
Software
Engineering
CPU
1st
Step:
The Generalization of
Software Engineering
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 12
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010, 67
Programming Language Paradigms
(1)
language category Computer Languages Languages f. Anti Machine
both deterministic procedural sequencing: traceable, checkpointable
operation sequence driven by:
read next instruction, goto (instr. addr.),
jump (to instr. addr.), instr. loop, loop nesting
no parallel loops, escapes, instruction stream branching
read next data item, goto (data addr.),
jump (to data addr.), data loop, loop nesting, parallel loops, escapes, data stream branching
state register program counter data counter(s)
address computation
massive memory cycle overhead overhead avoided
Instruction fetch memory cycle overhead overhead avoided
parallel memory bank access interleaving only no restrictions
language features control flow + data manipulation
data streams only (no data manipulation)
Flowware Languages Software Languages
imperative language twins
© 2008, [email protected] http://hartenstein.de 2010, 2010, 68
Programming Language Paradigms
(2)
Computer Languages Languages f. Anti Machine
procedural sequencing: traceable, checkpointable
read next instruction, goto (instr. addr.),
jump (to instr. addr.), instr. loop, loop nesting
no parallel loops, escapes, instruction stream branching
read next data item, goto (data addr.),
jump (to data addr.), data loop, loop nesting, parallel loops, escapes, data stream branching
program counter data counter(s)
massive memory cycle overhead overhead avoided
memory cycle overhead overhead avoided
interleaving only no restrictions
control flow + data manipulation
data streams only (no data manipulation)
Flowware Languages Software Languages
imperative language twins
© 2008, [email protected] http://hartenstein.de 2010, 2010,
The Machine Model Dichotomy
(2) von Neumann versus data stream machine
69 PE
Program Engineering
Flowware
Engineering
FE
auto-sequencing Memory
asM
SE
Software
Engineering
CPU
CE
Configware Engineering structures
pipe network
model etc. DPU Data-Path- Unit
Data-Path- Array DPA
The Generalization of
Software Engineering —
2nd
Step:
© 2008, [email protected] http://hartenstein.de 2010, 2010,
The Machine Model Dichotomy
(3) von Neumann versus data stream machine
70 PE
Program Engineering
Flowware
Engineering
FE
auto-sequencing Memory
asM
SE
Software
Engineering
CPU
CE
Configware Engineering structures
pipe network
model etc. DPU Data-Path- Unit
Data-Path- Array DPA
time to time time to space
mapping issue
The Generalization of
Software Engineering —
2nd
Step:
© 2008, [email protected] http://hartenstein.de 2010, 2010, 71
A Clean Terminology, please
program source result
Software instruction streams
Flowware data streams
Configware datapath structures configured
© 2008, [email protected] http://hartenstein.de 2010,
CPU-centric Flat World Model
Aristotelian model
This Software-only world model
is obsolete CPU not visible from SE
(CS: introduced in the 40ies)
72
CPU-centric sequential-only
mind set
CPU-centric sequential-only
mind set
1,000,000 100,000 10,000
1000 100 10
but no hardware know-how
but no hardware know-how
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 13
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Outline
• The coming Shortage of Energy
• Why Computing is important
• Energy Consumption of Computing
• The Programmability Crisis
• Rescue by Reconfigurable Computing ?
• The Reconfigurability Paradox
• We need to Reinvent Computing
• Reinventing Programmer Education
• Conclusions
73 © 2008, [email protected] http://hartenstein.de 2010, 2010,
The Language and Tool Disaster
Software people don‘t speak VHDL
Hardware people don‘t MPI nor OpenMP
Bad quality application development tools
86% designers hate their tools [FCCM’98]
progress stalled by qualification problems in industry and academia
Comprehensibility barrier between procedural and structural mind set
N. Conner et al.: FPGAs for Dummies; Wiley, 2008
74
© 2008, [email protected] http://hartenstein.de 2010, 75
Dual Paradigm Mind Set: an old Hat
Mapped into a Hardware mind set: action box = Flipflop, decision box = (de)multiplexer
Software mind set: instruction-stream-based: flow chart -> control instructions
(mapping from procedural to structural domain)
C. G. Bell et al: The Description and Use of Register-Transfer Modules (RTM's); IEEE Trans-C21/5, May 1972
W. A. Clark: Macromodular Computer Systems; 1967 SJCC, AFIPS Conf. Proc. 1967:
1972:
FF
token bit
evoke
FF FF
We need
© 2008, [email protected] http://hartenstein.de 2010, 2010,
POIIP: Loop turns into Pipeline
76
[1979]
(reconfigurable) DataPath Unit:
rDPU loop body
rDPU
rDPU
rDPU
Pipeline:
rDPU loop body
loop:
complex loop body nested loops
complex rDPU or pipe network inside rDPU
complex pipe network
CPU
Memory
© 2008, [email protected] http://hartenstein.de 2010, 2010, 77
Locality Awareness is
essential for Flowware
How data are moved
Software: by addresses, read from instruction
Flowware: by wire (configured before run time)
time to space mapping
Software
Configware
Flowware
space to time mapping
locality awareness
Programmer instruction for Reconfigurable Computing:
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Development with VHDL is expensive
78
FPGAs Achilles’ Heel is their long development time: Low level HDLs (VHDL/Verilog) are still dominant
Unlike software, FPGAs do not offer forward/backward compatibility
[Khaled Benkrid, Un. of Edinburg] FPGAs: low technology maturity; small user base, compared to software”
FPGA ASIC eASIC
Variety of Tool Flows
Survey urgently needed: Manpower required!
Complicated IP Core Scene
BDTI High-Level Synthesis
Tool Certification Program™ Grant Martin, Gary Smith. “High-Level Synthesis: Past, Present, and
Future,” IEEE Design and Test of Computers, July/August 2009.
TSMC creates a 'soft' IP cores collaboration program to improve soft IP readiness for EDA and IP suppliers including Arteris, Atrenta, Cadence, Chips & Media, Imagination Technologies, Intrinsic-ID, MIPS Technologies, Sonics, Synopsys and Vivante.
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 14
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Productivity vs. Efficiency
79
[courtesy Richard Newton]
„The nroff of EDA“ [R. N.]
how to hide the ugliness from the user [Herman Schmit]
HDLs: zero ease of use !!!
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Language-of the Year Phenomenon
[R. Newton]
[courtesy Richard Newton]
80
KARL
© 2008, [email protected] http://hartenstein.de 2010, 2010,
HLL programming models
81 © 2008, [email protected] http://hartenstein.de 2010, 2010,
Understanding Complex
Hetero Systems
82
Layers of Abstraction and Automatic Parallelization hide critical sources of, and limits to efficient parallel execution
Efficient Distribution of Tasks being memory limited
Internode Communications reduces Computational Efficiency
We must change how programmers think
essential: awareness of locality,
Focusing on memory mapping issues and transfer modes to detect overhead and bottlenecks
Understanding streams through complex fabrics needed
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Understanding Architecture
83
NoC
memory
memory
memory
ASIC
ASIC
ASIC
ASIP
ASIP
ASIP
FPGA
FPGA
FPGA
µP
µP
µP
I/O
I/O
I/O
ANN
ANN
memory,
streams
off-chip
th
e m
em
ory
wa
ll
several transfer modes
reconfigurable accelerators
hardwired accelerators
self-reconfigurable accelerators
many-core
© 2008, [email protected] http://hartenstein.de 2010, 2010,
>60 NoC Projects
Jih-Sheng Shen, Pao-Ann Hsiung (editors): Dynamic Reconfigurable Network-on-Chip Designs: Innovations for Computational Processing and Communication; Information Science Reference, Hershey, USA, April 1, 2010
84
Industry faces 'platform collision'
Which platform technology will win ? ASIC, ASSP, FPGA, MCU or IP core?
NoC research: world-wide >60 projects
"It's not clear, all may coexist”
Battles will get further interesting if/when the parallel programming crisis is over
[Brad Howe,
VP IC, Altera]
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 15
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Variety of Architectures
85
AS Application-Specific DP DataPath DPA DataPath Array EPP Extensible Programmable
Platform FP field-programmable GA Gate Array IC Integrated Curcuit IP Instruction Set Processor PU Processing Unit R reconfigurable
CPU, IP
FPDPA, FPGA
FPASIP
(Tensilica, Fujitsu)
ASIC
EPP (Xilinx)
ASIP, ASPU
U Unit
…. and
many,
many
more
X86, ARM … Temsilica
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Parallel Architecture Visualization
86
Pipe Network Model
RTL Visualization
Supporting Locality Awareness
EDA
C++ ? C
IP Cores
other Models? (concurrent)
new mix of skills needed
essential: locality awareness
focus on memory mapping issues and transfer modes to detect
overhead and bottlenecks
understanding streams through complex fabrics
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Newer Developments in
Semiconductor Technology
Limits by increasing power density
87
significant problems in performance, power consumption and reliability: great challenges for Reconfigurable Computing.
the golden
CMOS era
is gone
Technology scaling does not deliver significant performance speedup
transistors less reliable: additional sources of errors*
defective at manufacture time
degrade and fail over the expected lifetime
process variations
increasing number of soft errors
December 28, 2009
Fault-Tolerance Techniques needed
(EM, HCD, TDDB)
*) J. M. P. Cardoso, M. Hübner (editors): Reconfigurable Computing, 2011, Springer *) S. Borkar: Designing reliable systems from unreliable components; 2005. © 2008, [email protected] http://hartenstein.de 2010, 2010,
Fault Tolerance
88
CPU
hardwired
accelerators
reconfigurable
accelerators
Fault Tolerance Implementation
accelerators
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Neurocomputing
89
The Memristor
hp:
discovered
2008, prod.
announced
for 2013
direct synapse emulation replacing massively inefficient digital simulation
less transistors for logic circuits
resistor w. memory: doping moved by electric field
third paradigm:
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Mem(r)istor History
1963: Memistor Corp. founded by Prof. Bernie Widrow, Stanford U.
Foto: Storz 1975
1960: missing device postulated by Prof. Karl Steinbuch
2007: Stan Williams, hp, finds Memristor
[Picture: Leon Chua]
1971: Leon Chua, UCB, specifies Memristor
90
1958: Mark I Perceptron: 512 motor-driven potentiometers
2013 ? agreement Hewlett Packard,
Hynix Semiconductor
Frank Rosenblatt
Lernmatrix: ANN patents in the 50ies
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 16
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Triple paradigm systems?
91
accelerators CPU
hardwired
accelerators
reconfigurable
accelerators
(self-reconfigurable) neurocomputing
Self-organizing Fault Tolerance
© 2008, [email protected] http://hartenstein.de 2010, 92
[Olivier Temam: The Rebirth of Neural Networks; 37th ISCA, June 19-23, 2010, Saint-Malo, France]
[Olivier Temam, 2010]
[Olivier Temam,
2010]
What ANNs can do
[picture: Olivier Temam, 2010]
1989: single hidden layer enough
© 2008, [email protected] http://hartenstein.de 2010,
Defects-Tolerant Accelerators ?
93 [Olivier Temam, 2010]
for self-reconfigurable neuro computing
© 2008, [email protected] http://hartenstein.de 2010, 2010, 94
Reconfigurable
Neuro Computing or: x1 + x2 + x3 ≥ 1
and: x1 + x2 + x3 ≥ 3
w. Memristor LUT
VDD
RNC
from Boolean Algebra
Generalization of the LUT
to Steinbuch Algebra
from Reconfigurable Computing to Reconfigurable Neuro
Computing (RNC)
Field-Programmable
Neuron Array (FPNA)
Threshold Logic (Majority Logic)
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Outline
• The coming Shortage of Energy
• Why Computing is important
• Energy Consumption of Computing
• The Programmability Crisis
• Rescue by Reconfigurable Computing ?
• The Reconfigurability Paradox
• We need to Reinvent Computing
• Reinventing Programmer Education
• Conclusions
95 © 2008, [email protected] http://hartenstein.de 2010, 2010, 96
Conclusions
Legacy Software creates Masses of Jobs for Decades ….
To avoid future unaffordability of our Cyber Infrastructure we need a Massive Migration Effort
achieves much more than most Climate Protection Proposals
RC is the Silver Bullet
We have to activate the Public and the Media, currently fully ignoring this wordwide vital issue
We must start this Campaign as long as we can afford it
A fascinating Challenge to reach new Horizons of research.
We need a Twin Paradigm Co-Education
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 17
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010, 97
http://hartenstein.de/bio.pdf
Obrigado!
http://hartenstein.de/BGME-10.pdf
© 2008, [email protected] http://hartenstein.de 2010, 2010, 98
END
© 2008, [email protected] http://hartenstein.de 2010,
Backup for Discussion
99 © 2008, [email protected] http://hartenstein.de 2010, 2010,
Rewriting needed anyway
• Rewriting of software needed anyway: for the survival ot the µP industry (to cope with the transision to manycore)
100
• Extended scope of Software Rewriting: to save energy by orders of magnitude
• different from „classical“ green computing
stro
ng s
yne
rgy
© 2008, [email protected] http://hartenstein.de 2010,
ANN 101 © 2008, [email protected] http://hartenstein.de 2010,
normal Hype Curve
102
[Olivier Temam: The Rebirth of Neural Networks; 37th ISCA, June 19-23, 2010, Saint-Malo, France]
[Olivier Temam, 2010]
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 18
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010,
normal Hype Curve
103
[Olivier Temam: The Rebirth of Neural Networks; 37th ISCA, June 19-23, 2010, Saint-Malo, France]
[Olivier Temam, 2010]
© 2008, [email protected] http://hartenstein.de 2010,
Neural Network Hype Curve
104
[Olivier Temam: The Rebirth of Neural Networks; 37th ISCA, June 19-23, 2010, Saint-Malo, France]
[Olivier Temam, 2010]
(Olivier Temam never mentions Karl Steinbuch)
© 2008, [email protected] http://hartenstein.de 2010,
Neural Network Hype Curve
105
[Olivier Temam: The Rebirth of Neural Networks; 37th ISCA, June 19-23, 2010, Saint-Malo, France]
[Olivier Temam, 2010]
(Olivier Temam never mentions Karl Steinbuch)
© 2008, [email protected] http://hartenstein.de 2010,
no
stopped Funding for 15 Years
106
[Olivier Temam,
2010]
Marvin Minsky,
Seymour Papert:
Perceptrons; 1969.
(world-wide)
SVM: Support vector machines: set of supervised learning methods to analyze data and recognize patterns
© 2008, [email protected] http://hartenstein.de 2010,
no
Marvyn Minski‘s blind alarm
107
20 y
ears
earl
ier
!
W. Hilberg: Karl Steinbuch, ein zu Unrecht vergessener Pionier der Künstlichen Neuronalen Systeme; FREQUENZ 1995, 49#(1-2):28-35.
[Olivier Temam,
2010]
1962: Karl Steinbuch
(was ignored by Marvyn
Minski book)
Steinbuchs Lernmatrix
1960
© 2008, [email protected] http://hartenstein.de 2010,
miscellaneous
108
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 19
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Complex Programmability Crisis
109
Concurrent vs. FPGA Programming
hetero Programming
Fine Grain vs. Coarse Grain Programming vs.
Survey urgently needed: Manpower required!
Acceleration needs an in-depth Application Study: requires more Programming Effort
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Processor inside FPGA vs.
FPGA inside Processor: EPP
110
-> totally changed concept
device more like heterogeneous SoCs: significant benefits for HPC applications:
FPGAs became
software-centric:
-> EDUCATION !!!!-
not hardware centric:
Xilinx: Extensible Processing PlatformTM
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Structured ASICs
Structured ASICs like eASIC are based mostly on FPGA-like architecture with special configuration mechanism to program at mask level: not re-programmable (more performance for less cost).
111
configware on ROM
© 2008, [email protected] http://hartenstein.de 2010, 2010,
RTL Programming to ASIC / ASSP
Platform-Language of Silicon: attractive for IP providers
112
Tools-Path to ASICs/ASSPs, across FPGAs
RTL is inherently parallel, mapped application are automatically optimally parallelized by CAD tools.
Application-Specific Standard Products
now ESL (Electronic System Level) bridging HDL and ANSI C/C++ (at industrial level )
Battle between FPGAs vs MPSoC:
it is RTL vs Software programming.
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Design methods and tools
Tools have impact on designer productivity
113
Synthesis tools focus on:
automatically mapping high level descriptions into efficient hardware implementations
according to performance metrics, such as speed, size, and power consumption.
can support verification of functionality, timing and
testability
covers all the main steps of synthesis and analysis, including:
capturing domain-specific knowledge,
profiling,
design space exploration,
multi-core partitioning,
(H/S)system partitioning,
data representation optimization,
static and dynamic reconfiguration,
optimal custom instruction set generation,
functional simulation,
programming mixed SW/RHsystems,
ensuring effective cross-boundary communication (another challenging issue).
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Graphical / Dataflow Languages
114
E. El-Araby et al.: Comparative Analysis of High Level Programming for Reconfigurable Computers: Methodology And Empirical Study; Proc. SPL2007 Symp., Mar del Plata, Argentina, Febr. 2007
We should have a look at DSPlogic
We should have a look at DSPlogic
Evaluation Metrics
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 20
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
New boundary constraints are the limiting factor
115
Legacy scientific applications: predominantly sequential
The entire software ecosystem needs to evolve (including curricula): O/S, Libraries SW development environments, compilers and languages
additional levels of parallelism: chaining, pipelining, systolic, super-systolic, wavefront arrays additional data structures and
storage organization: the new distributed memory discipline
Evolve the Software Ecosystem
© 2008, [email protected] http://hartenstein.de 2010,
Tools for Team Design
28-nm FPGAs deliver the equivalent of a 20- to 30-million gate ASIC.
Tools for team-design are coming up:
• Distributed and parallel development,
• Design flows,
• Tracking and reporting
Top level subproject management and integration, including
• source code version control
• Design / IP re-use
• Time budgeting
• In-context synthesis
116
FPGA design tool operation in a reasonable time impossible
© 2008, [email protected] http://hartenstein.de 2010, 2010,
>15000
PISA project
117
FFT
100
Reed-Solomon Decoding 2400
Viterbi Decoding 400
1000
MAC
DSP and wireless
molecular dynamics simulation
88
BLAST 52
protein identification
40
Smith-Waterman pattern matching
288
Bioinformatics
GRAPE
20 Astrophysics
SPIHT wavelet-based image compression 457
real-time face detection
6000
video-rate stereo vision
900
pattern recognition 730
Image processing, Pattern matching, Multimedia
3000 CT imaging crypto
1000
28500
DES breaking
1
1000
1,000,000
Spe
edup
-Fac
tor
Speed-up
factors are
not new
by avoiding the
von Neumann
paradigm
8723 DNA seq.
100
10
10,000
100,000
© 2008, [email protected] http://hartenstein.de 2010,
118 Speed-up the mid’ 80ies
PISA project
one DPLA* replacing 256 early Xilinx FPGAs
*) fabricated by E.I.S. project
© 2008, [email protected] http://hartenstein.de 2010,
The transition from machine level to higher level languages led to the biggest productivity gain ever made
It‘s alarming that today‘s megabytes of code are compiled from languages at low abstraction levels (C, C++,Java)
The wrong Direction: by Herd Instinct ?
[Fred Brooks]
119
Java is a religion – not a language [Yale Patt]
Bud Lawson‘s Dilbert Cartoon on the Process Manager
„portable assembler language“
© 2008, [email protected] http://hartenstein.de 2010,
Burroughs B5000/5500: language-friendly stack machine
IBM 260/370 & intel x86 highly complex instruction set
MULTICS (GE, Honeywell): well manageable (impl. in PL/1)
UNIX: complexity problems, compatibility problems
Pascal killed by C, coming as an infection, along with UNIX
unnecessary complexity
inside
Widening the Semantic Gap
[Harold „Bud“ Lawson]
„portable assembler language“
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 21
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010, 121
Danger of the Situation
Unnecessary Complexity
1963 2010
© 2008, [email protected] http://hartenstein.de 2010, 2010, 122
Danger of the Situation
Unnecessary Complexity
1963 2010
Parallel Programming Gap
© 2008, [email protected] http://hartenstein.de 2010, 2010, 123
Scientific Revolutions
scientific scenes follow the herd instinct
Thomas Kuhn
1st Newtons Law (inertia): „people do not change direction“
Gordon Bell
© 2008, [email protected] http://hartenstein.de 2010, 2010,
>40 years Software Crisis
F. L. Bauer 1968, coined „Software Crisis“ - N. N. 1995: THE STANDISH GROUP REPORT Robert N. Charette 2005: Why Software Fails; IEEE Spectrum, Sep 2005
Anthony Berglas 2008: Why it is Important that Software Projects Fail
Oct 1957
The Economist: Nov 19th 1955
In 1955, Parkinson could not have foreseen the impact of software.
The size of bureaucracy is independent of the amount of real work to be done.
124
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Apropos Herd Instinct
125
Some Programming Languages
G. Bell: „People do not change Direction“ (follow Herd Instinct)
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Architecture visible
The software programming model:
Which hardware architecture parts are visible
and under the programmer’s direct control ?
126
The RC programming model:
how the programmer can control data transfers between FPGA, onboard memory, the microprocessor and its memory
to the programmer
http://hartenstein.de
Keynote, The 1st Brazilian-German Workshop on Micro and Nano Electronics (BGME’2010), October 6 - 8, 2010, Porto Allegre, RS, Brazil 22
Reiner Hartenstein. Informatik Department, TU Kaiserslautern, Germany
© 2008, [email protected] http://hartenstein.de 2010, 2010,
The Human Brain …
“The human brain needs only a few watts for about 10,000,000,000,000,000 (ten million billion) compute operations per second.
127
Von-Neumann-paradigm-based it would need more than a MegaWatt, the electric power needed for a small town and it would sizzle off in a fraction of a second.” [Alfred Fuchs].
© 2008, [email protected] http://hartenstein.de 2010, 2010,
SoC - System-on-Chip
128
System on a Chip development:
Yesterday’s chip is today’s functional block!
New design methodologies needed
3.0 1.0 0.5 0.2
20K gates
Schematics
& simulation
50K gates
Schematics
& Synthesis
500k gates
Simulation,
Emulation,
Synthesis,
Formal equivalence
2.5 million gates New Design Paradigm
Source: ICE
[Courtesy Manfred Glesner]
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Vertical Disintegration
[Courtesy Manfred Glesner]
129
Electronic
Company
Foundry Foundry
1960 200X
IP
Company Semicon-
ductor
company
Fabless
Semicon-
ductor
Fabless
Semicon-
ductor
System
company
System
company
System
company
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Participants in the Value Chain
130
Source: Gartner
[Courtesy Manfred Glesner]
© 2008, [email protected] http://hartenstein.de 2010, 2010,
Twin paradigm systems
131
CPU „Central Processing Unit“
„Central“: it controls
(almost) everything
However,
it needs
accelerators
CPU
hardwired
accelerators
reconfigurable
accelerators second paradigm
accelerators
ASIC: 3%
FPGA: 97%
[Dataquest, 2009]
design start ratio
© 2008, [email protected] http://hartenstein.de 2010,
Neural Network Hype Curve
132
[Olivier Temam: The Rebirth of Neural Networks; 37th ISCA, June 19-23, 2010, Saint-Malo, France]
[Olivier Temam, 2010]
(Olivier Temam never mentions Karl Steinbuch)