David E. Culler & Costas Spanos University of California, Berkeley March 14, 2011.

Welcome

to EECS Grad Student Visit Day 2011

David E. Culler & Costas SpanosUniversity of California, BerkeleyMarch 14, 2011

Why do graduate study in EECS @ UC Berkeley?

2

You are here!

3

… in Berkeley“Berkeley – the Athens of the West – is arguably the

world’s best place to live.” New York Times

Fabulous restaurants, theatre, parks, scenery, weather Culture: a community of independent thought, nonconformity

“The campus of the University of California at Berkeley in springtime is about as close to Shangri-La as most mortals are likely to get.”

New York Times

4

@ University of California, BerkeleyTimes Higher Education Supplement Worldwide University Ranking

Academic Reputation1. Berkeley2. Harvard

National Research Council Survey of Graduate Programs

1. Berkeley: 97% depts top 10 2. MIT/Harvard: 87%

5

…at the dawn of a new AgeGraduation

Window

You are here!

6

… of integration across vast scale

years

ComputersPer Person

103:1

1:106

Laptop

PDA

Mainframe

Mini

Workstation

PC

Cell

1:1

1:103

Mote!

Bell’s Law: new computer class per 10 years7

… in a changing World

8

1969

2.0 B 1/26/11

1974

RF

C 6

75 T

CP

/IPWWW

AR

PA

Net Internet

HT

TP

0.9

1990 2010

Why do graduate study in EECS @ UC Berkeley?

9

Why? - Facilities

Soda Hall

Cory Hall

Sutardja-Dai Hall

10

Microlab – 45 years of commitment to the future

11

Marvell Nanolab and Sutardja-Dai CITRIS Headquarters inaugurated 2/27/2009

12

The best Academic Cleanroom in the World

13

Why? - Academic Reputation

Times Higher Education SupplementWorldwide University Rankings, 2004-2008

Engineering and Information Technology

1. Berkeley

2. MIT

3. Stanford

Overall Academic Reputation

1. Berkeley

2. Harvard

National Research Council*Overall Academic Quality

1. Berkeley (35/36 departments in top 10)

2. Stanford (31/36)

3. Harvard (26/36)

National Science FoundationFellows Chosen Institution 2009

1. Berkeley (103 Fellows)

2. Stanford (58 Fellows)

US News & World ReportUS Graduate School Rankings, 2010

Computer Science Programs1. Berkeley/MIT/Stanford (3-way tie)

Computer Engineering Programs1. Berkeley/MIT/Stanford (3-way tie)

EE/Electronics Programs1. Berkeley/MIT/Stanford (3-way tie)

* recent numerical rankings are “work in progress” 14

15

Why? – Distinguished Faculty…

National Medal of Science (2) ACM A.M.Turing Award (3) MacArthur Prize (3) National Academy of Sciences (10) National Academy of Engineering (38) IEEE Medal of Honor (3) SIAM von Neumann Lecture Prize (2) American Society for Engineering Education

Awards (8) C&C Promotion Prize (2) Silicon Valley Engineering Hall of Fame (2) Benjamin Franklin Medal (3) Harvey Prize (1) Honda Prize (1) Kyoto Prize (1) Okawa Prize (2) National Science Foundation Awards (52) American Academy of Arts & Sciences Fellows (15) UC Berkeley Distinguished Teaching Award (12) Sloan Foundation Fellowships (11) ACM Doctoral Dissertation Award (12) Endowed Chairs (21) Many other ACM, IEEE, SIAM and other awards

The Faculty Teach, Advise and Lead Research:

• ~120 Lecture hours per Year• ~20 undergraduate advisees• ~6 graduate students• ~$600K in funded research• ~2 departmental committees• ~1 college or campus

committee

Nearly every faculty contributes to a World Class research activity.

Why? – New Faculty

Dr. Sylvia Ratnasamy 7/2011

• Area: Networked Systems co-inventor of Distributed Hash Tables Scalable software routers Theoretical formulation of “protocol simplicity” Energy efficiency in networked systems Internet architecture and protocols

Degrees: B.E. from University of Pune, 1997 Ph.D. Berkeley, 2002 (Shenker and Stoica)

Topic: “A Scalable Content-Addressable Network”

Positions: Research Staff, ICSI Center for Internet Research

(1999-2002) Senior Researcher, Intel Research Berkeley

(2002-present)

Dr. Ana Arias 1/2011 Area: Physical Electronics - Printed Organic

Electronics Processed electronic materials for flexible sensors Fully printed blast dosimeter Correlation of deposition conditions and

morphology on device performance Solving the “coffee ring effect

Degrees: B.S. and M.S. in physics, Federal University of

Paraná, Brazil (1995, 1997). Certificate in Physics Teaching (1995)

Ph.D. in physics, University of Cambridge, 2001 (Richard H. Friend) Topic: “Conjugated Polymer Blends Phase

Separation and Three-Dimensional Thin-Film Structure for Photovoltaics”

Positions: Group Leader/Engineer, Plastic Logic Limited

(2001-2003) Printed Electronic Devices Area Manager, Palo

Alto Research Center (2003-present)16

17

Why? Strong Interactions With Industry

Culture of Use-Inspired Fundamental Research

Proximity to Silicon Valley Strong industrial funding for research Internships Many startups

Create Industries, not just companies

NRC CSTB 2004 – Tracks to B$ Market

What do our competitors say?

2010 External Review Committee Report “If Berkeley is arguably the crown jewel of

America’s research-intensive universities, then EECS is arguably the crown jewel of Berkeley.”

“It is unsurpassed as a training ground both for the next generation of scholars and for the next generation of practitioners.”

“The research programs in Berkeley EECS are arguably the best in the world. Over many decades, Berkeley EECS has consistently opened up new areas of research for others.”

19

Why do graduate study in EECS @ UC Berkeley?

20

Because Berkeley EECS grad students change the world

SPICE – Simulation Program with Integrated Circuits Emphasis

21

70 80 90 00 10 20

A. Richard Newton

EE 223 (F'69), EE 225A (W'70), EE 225B (S'70)

Nagel, L. W, and Pederson, D. O., SPICE (Simulation Program with Integrated Circuit Emphasis), Memorandum No. ERL-M382, University of California, Berkeley, Apr. 1973

RISC – Reduced Instruction Set Computers

22

70 80 90 00 10 20

M. Katevenis, R. Sherburne, D. Patterson and C. Sequin: ``The RISC II Micro-Architecture'', Proceedings of VLSI '83, Trondheim, Norway, Aug. 1983

Manolis Katevenis

23

Research as “Time Travel” - the secret formula Imagine a technologically plausible future Create an approximation of that vision

using technology that exists. Discover what is True in that world

Empirical experience Bashing your head, stubbing your toe, reaching

epiphany Quantitative measurement and analysis Analytics and Foundations

Courage to ‘break trail’ and discipline to do the hard science

NOW – Scalable High Performance Clusters

24

10th ANNIVERSARY REUNIONSNetwork of Workstations (NOW): 1993-98

25

NOW Team 2008: L-R, front row: Prof. Tom Anderson†‡ (Washington), Prof. Rich Martin‡ (Rutgers), Prof. David Culler*†‡ (Berkeley), Prof. David Patterson*† (Berkeley). Middle row: Eric Anderson (HP Labs), Prof. Mike Dahlin†‡ (Texas), Prof. Armando Fox‡ (Berkeley), Drew Roselli (Microsoft), Prof. Andrea Arpaci-Dusseau‡ (Wisconsin), Lok Liu, Joe Hsu. Last row: Prof. Matt Welsh‡ (Harvard/Google), Eric Fraser, Chad Yoshikawa, Prof. Eric Brewer*†‡ (Berkeley), Prof. Jeanna Neefe Matthews (Clarkson), Prof. Amin Vahdat‡ (UCSD), Prof. Remzi Arpaci-Dusseau (Wisconsin), Prof. Steve Lumetta (Illinois).

*3 NAE members †4 ACM fellows ‡ 9 NSF CAREER Awards

Inktomi – Fast Massive Web SearchFiat Lux - High Dynamic Range Imaging

26

70 80 90 00 10 20

Paul Gauthier

Paul Debevec

Fiat Lux

Research in Courses First-year courses such as CS262AB

(systems), CS281AB (learning), EE227AB (convex optimization) include research projects Learn to do research in a stress-free way Multiple courses => interdisciplinary research! Term projects (almost) conference pub

20-30 “special topics” courses every year focus on cutting edge research areas RISC, RAID, NOW, SPICE all started in advanced

graduate courses Launch successful projects, 1st or 2nd area of

expertise27

Computational Lens on the Sciences

28

70 80 90 00 10 20

Costis Daskalakis

Constantinos Daskalakis, Paul W. Goldberg and Christos H. Papadimitriou, The Complexity of Computing a Nash Equilibrium, In the 38th ACM Symposium on Theory of Computing, STOC 2006

Berkeley EECS Research Style

Work together on really important problems No matter how hard

Have ideas with impact

Produce great students

29

Research Styles: Collaboration

Advisor and student working together Most meet at least weekly

Advisor and several students working together

Several faculty and many students Sensor Networks, ParLab, Nanolab, BWRC,

BSAC Groups of students with faculty guidance

E.g., Probabilistically Checkable Proofs - PCP

30

A Day at Berkeley

Professors Spanos and Poolla and their typical research group meeting.

31

Research Retreats Project Reviews with Outsides

Twice a year: 3-day retreat at nice place

Faculty, students, staff, industry Industry visitors supply feedback Faculty members listen to it! High-intensity informal exchanges Builds team spirit

Experience shows that research retreats are the key ingredient in the success of large-scale (10-25 person) projects

32

EECS

Harnessing physical processes to perform logically defined functions Everything from band-gap phenomena to

Google and Avatar

Dense interconnect between EE and CS, and increasingly between EECS and statistics

33

Research

34

Some of Our Contributions:

Berkeley UnixThe first free Unix, virtual memory, foundation of Linux

Computational complexityNP-completeness

CryptographyFoundations of cryptographic protocols

Approximation hardnessPCP (Probabilistically Checkable Proofs)

DevicesFINFET transistor, organic semiconductors, etc.

Electrical ground fault interruptorsInvented at Berkeley in the 1950s

Electronic design automation (EDA)Berkeley built this industry

Embedded systemsConcurrency, real-time computing, formal foundations

Floating pointIEEE 754 floating point standard

Graph algorithmsNetwork Flow, Planar separators and embeddings

Hybrid systemsMixed discrete/continuous systems

Nanoscale electronicsPhotolithography, transistors, transistor models, etc.

NetworkingTCP/IP, foundation of the Internet, in Berkeley UNIX

Mixed-signal circuitsKey contributions that make CMOS dominant

• MEMS systemsmicroelectromechanical systems

• Model-based designConcurrent models of computation, formal foundations

• Modern probabilistic AIReunified AI, learning, vision, control theory, stats

• Open source movementBerkeley software is truly free (vs. MIT’s GPL)

• Quantum computingFoundations of Quantum Complexity Theory

• Parallel computingNetwork of Workstations

• RAID storage systemsDominant design for large storage systems

• Randomized algorithmsRandomness as a computational resource

• Relational databasesAn EE/CS collaboration (Stonebraker & Wong)

• RISC processorsReduced instruction set computers

• Sensor NetworksBerkeley created this field

• Soft computingFuzzy logic

• Systems theoryFoundations of control, communications, signal proc.

• SpiceWorldwide standard in circuit simulation

35

Our General Approach to the PhD

We want you to acquire great minds Deep, broad, skilled in the art of creation Much further beyond your current self than you are

beyond your high-school self

It’s important to learn new ways of thinking Multidisciplinary research projects Courses in related areas of EE and/or CS Courses in related disciplines: statistics, mathematics

(geometry, logic, …), molecular biology, materials science, quantum physics, …

We want you to have a good time doing it (Ask the current students about this part.)

Our approach seems to be working

36

Berkeley PhDs in top-15 EE deptsCal TechBridges, William Doyle, JohnLow, StevenMurray, Richard M.Perona, PietroRutledge, David Tai, Yu-ChongYariv, Amnon

USCBreuer, MelvinDimakis, AlexFeinberg, JackHwang, KaiKim, Eun SokPedram, Massoud

UCLAAbidi, AsadCabric, DanijelaChiou, Pei YuDolecek, Lara Jacobsen, Stephen E.Judy, JackMarkovic, Dejan Srivastava, ManiWang, Paul K.C

University of MichiganKu, Pei-Cheng Lafortune, StéphaneMaharbiz, MichelNguyen, XuanLongPradhan, SandeepTilbury, DawnZhang, Zhengya

BerkeleyAnantharam, VenkatBahai, Ahmad Budinger, ThomasChang-Hasnain, ConnieHu, ChenmingLee, EdwardMorgan, Nelson Newton, RichardNguyen, Clark Niknejad, AliPister, KristoferPolak, Lucien* Sastry, Shankar Shank, CharlesTomlin, ClaireVan Duzer, Theodore* Varaiya, Pravin* Walrand, JeanWelch, William J.* Whinnery, John*Wu, Ming

StanfordBambos, NickBoyd, Stephen P.De Micheli, GiovanniDutton, Robert W.Goldsmith, AndreaHowe, RogerKahn, JosephMcKeown, NickMeng, TheresaMurmann, BorisPoon, Ada Wong, S. SimonWooley, Bruce

UIUCAdesida, IlesanmiCangellaris, AndreasChiu, YunDeTemple, ThomasGross, GeorgeHajek, BruceHajj, IbrahimKwiat, PaulMa, Yi Pai, AnanthaRosenbaum, Elyse Viswanath, Pramod

PurdueHu, JianghaiSands, Timothy

UT AustinArapostathis, AristotleBaldick, Ross Chen, Ray de Veciana, Gustavo Garg, VijayGharpurey, RanjitLee, JackOrshansky, Michael

CornellAvestimehr, SalmanMolnar, AlyoshaWagner, Aaron

MITChandrakasan, AnanthaColeman, Charles P. Daniel, Luca Goyal, Vivek Ippen, ErichLee, Hae-SeungSodini, CharlesWhite, JacobZheng, Lizhong

HarvardWood, Robert

CMUFedder, GaryRohrer, Ronald

University of MarylandAbed, EyadBhattacharyya,ShuvraGligor, VirgilLa, Richard Newcomb, RobertRosfjord, Kristine Tits, AndreZaki, Kawthar

Georgia TechBuck, JohnKornegay, Kevin Madisetti, VijayMay, GaryMilor, Linda

37

Berkeley PhDs in Top-16 CS DeptsUniv. of Washington

Borriello, GaetanoEggers, SusanGribble, StevenIvory, MelodyLadner, Richard Lee, JamesRuzzo, Walter

UC BerkeleyAsanovic, KrsteDemmel, James Garcia, DanielHearst, MartiKatz, RandyMcMains,Sara Paxson, Vern Song, Dawn Vazirani, UmeshWagner, David

Stanford UniversityGill, JohnHeer, Jeff Jurafsky, DanKlemmer, Scott Kozyrakis, Christos Levis, PhilMotwani, RajeevNg, AndrewRosenblum, Mendel

Cal TechPerona, PietroUmans, Christopher

UCLAKlinger, AllenMajumdar, RupakPotkonjak, MiodragTamir, Yuval

University of WisconsinArpaci-Dusseau, AndreaArpaci-Dusseau, RemziBach, EricCarey, Michael Chenney, Stephen Dewey, Colin Hill, MarkKlein, SheldonWood, David

University of MichiganChen, PeteMao, MorleySylvester, Dennis Newman, MarkDutta, Prabal

University of IllinoisBorisov, Nikita Erickson, JeffHwu, Wen-meiLumetta, StevenWah, BenYu, Yizho

USCLeonard Adleman

UT AustinArikan, Okan Dahlin, MikeDhillon, InderjitWarnow, TandyZuckerman, David  

PrincetonArora, Sanjeev  Blei, David M. Funkhouser, Tom

Carnegie MellonCooper, EricEfros, AlexeiFedder, GaryGibson, GarthGoldstein, SethGupta, AnupamHarchol-Balter, MorHeckbert, PaulHong, Jason Miller, GaryRohrer, Ronald Rudich, StevenSeshan, SriniXing, Eric P. Zhang, Hui 

MITBalakrishnan, HariDaskalakis, Costis Devadas, SrinivasGoldwasser, ShafiLampson, ButlerLiskov, BarbaraMadden, SamMicali, SilvioRubinfeld, RonittSipser, Michael Solar-Lezama, Armando Sudan, Madhu Teller, Seth

Harvard Grosz, BarbaraMitzenmacher, Mike Seltzer, MargoWelsh, Matt

Yale Angluin, DanaKrishnamurthy, Arvind

CornellBirman, KenShmoys, David

38

CS PhDs 1995-2005 in top 10 CS Depts

http://www.cs.wisc.edu/~estan/alumnistatistics/Alumni10_matrix.html

UCB MIT CMU SU0

5

10

15

20

25

30

With self-hires

Why? Because we care

AttendeesCornell 0CMU 2MIT 1Princeton 0Stanford 0UIUC 1UC Berkeley 34

40

The future of EECS

Radically new physics at the bottom: nano, quantum, negative capacitance…

Redesign the hardware stack, rethink levels of separation, retool the design industry

New machine architectures: massively parallel, probably stochastic, possibly quantum; and new ways to program them

Redesign the communication stack top-to-bottom: clean slate

Intelligent human-scale robots Information systems that know everything As-yet-unimagined forms of education,

entertainment Increasing extroversion: engaging with and

solving the problems of society

41

EECS program healthier than ever

Cory 4th floor Swarms + Post-Silicon + Photonics

Intel STC on Secure Computing Final round for Simon Institute for

Theoretical Computer Science …

42

EECS is more critical than ever

43

A different “Graduation Window”

Today

0°C

Feedback

Abrupt climate change

Water Rising seasWater shortagesGlaciers melt

Weather Storms, droughts, fires, heat waves

Ecosystems

Reefs damaged

Species extinction

Food Crop yields fall

3°C2°C1°C

Global temperature change (relative to pre-industrial era)

4°C 5°C

44

45

The worst thing about being at Berkeley is that you can never really be happy anywhere else - Prof. Shafi Goldwasser, MIT46

47

Why do graduate study in EECS @ UC Berkeley?

48

But, …

49

Berkeley’s financial situation Total campus spending increased in 2008-9 vs 2007-8

Research activity very healthy EBI ($500M over 10 yrs) is the largest grant to any university in history

Berkeley will have 70 faculty searches next year and following State support functions analogously to endowment, bridging gap

between tuition and the operating budget Tuition is rising, as is federal support

UC 20% drop in state support Equivalent to ~5-6% of campus budget In-state undergraduate tuition increased to ~ $11K (vs ~$38K peers) Proportion of out-of-state students will go from 8% to ~20% Actively diversifying income stream with professional masters, etc.

Stanford’s endowment lost 27% 9/08-8/09; 500+ layoffs, faculty hiring freeze

MIT’s endowment lost 21% 7/08-6/09; $125M budget cut, freeze

50

Didn’t the faculty have a pay cut?

51

% of salary covered by campus

52

% of salary covered by campus

53

% of salary covered by campus

54

The situation for PhD students PhD students don’t pay fees or depend on

state funds Bit more expensive for the faculty member

Campus fellowships for EECS more than doubled in 2010 Further increase in 2011

EECS research dollars up around 15% over 5 years to ~$65M/yr Not including new $25M NSF STC in low-power electronics

Most PhD students are paid 100% (i.e. double stipend) in summer

2005-6 2006-7 2007-8 2008-9 2009-100

10000

20000

30000

40000

50000

60000

70000

FederalNot for profitIndustryNon-Fed GovtUniv. of CA

55

Campus-wide trends since 2001

40% increase in PhD applicants Increase in offer acceptance from 45% to 55% 124 NSF fellows in 2009 (10% of US total)

110 at MIT, 72 at Stanford Berkeley has more Sloan Fellows than any other

institution, indicating quality of younger faculty $155M for PhD fellowships in current campaign

56

The bottom line

All major institutions took a hit; the appropriate response is collaboration - e.g., work on federal science policy

Universities are very robust and long-lived, whether private (Harvard, MIT, Stanford) or public (Berkeley, Oxford, Cambridge)

Make a choice on the scientific merits; the rest will take care of itself

57

Some of Our Contributions:

Berkeley UnixThe first free Unix, virtual memory, foundation of Linux

Computational complexityNP-completeness

CryptographyFoundations of cryptographic protocols

Approximation hardnessPCP (Probabilistically Checkable Proofs)

DevicesFINFET transistor, organic semiconductors, etc.

Electrical ground fault interruptorsInvented at Berkeley in the 1950s

Electronic design automation (EDA)Berkeley built this industry

Embedded systemsConcurrency, real-time computing, formal foundations

Floating pointIEEE 754 floating point standard

Graph algorithmsNetwork Flow, Planar separators and embeddings

Hybrid systemsMixed discrete/continuous systems

Nanoscale electronicsPhotolithography, transistors, transistor models, etc.

NetworkingTCP/IP, foundation of the Internet, in Berkeley UNIX

Mixed-signal circuitsKey contributions that make CMOS dominant

• MEMS systemsmicroelectromechanical systems

• Model-based designConcurrent models of computation, formal foundations

• Modern probabilistic AIReunified AI, learning, vision, control theory, stats

• Open source movementBerkeley software is truly free (vs. MIT’s GPL)

• Quantum computingFoundations of Quantum Complexity Theory

• Parallel computingNetwork of Workstations

• RAID storage systemsDominant design for large storage systems

• Randomized algorithmsRandomness as a computational resource

• Relational databasesAn EE/CS collaboration (Stonebraker & Wong)

• RISC processorsReduced instruction set computers

• Sensor NetworksBerkeley created this field

• Soft computingFuzzy logic

• Systems theoryFoundations of control, communications, signal proc.

• SpiceWorldwide standard in circuit simulation

58

Some of Our Contributions:

Berkeley UnixThe first free Unix, virtual memory, foundation of Linux

Computational complexityNP-completeness

CryptographyFoundations of cryptographic protocols

Approximation hardnessPCP (Probabilistically Checkable Proofs)

DevicesFINFET transistor, organic semiconductors, etc.

Electrical ground fault interruptorsInvented at Berkeley in the 1950s

Electronic design automation (EDA)Berkeley built this industry

Embedded systemsConcurrency, real-time computing, formal foundations

Floating pointIEEE 754 floating point standard

Graph algorithmsNetwork Flow, Planar separators and embeddings

Hybrid systemsMixed discrete/continuous systems

Nanoscale electronicsPhotolithography, transistors, transistor models, etc.

NetworkingTCP/IP, foundation of the Internet, in Berkeley UNIX

Mixed-signal circuitsKey contributions that make CMOS dominant

• MEMS systemsmicroelectromechanical systems

• Model-based designConcurrent models of computation, formal foundations

• Modern probabilistic AIReunified AI, learning, vision, control theory, stats

• Open source movementBerkeley software is truly free (vs. MIT’s GPL)

• Quantum computingFoundations of Quantum Complexity Theory

• Parallel computingNetwork of Workstations

• RAID storage systemsDominant design for large storage systems

• Randomized algorithmsRandomness as a computational resource

• Relational databasesAn EE/CS collaboration (Stonebraker & Wong)

• RISC processorsReduced instruction set computers

• Sensor NetworksBerkeley created this field

• Soft computingFuzzy logic

• Systems theoryFoundations of control, communications, signal proc.

• SpiceWorldwide standard in circuit simulation

59

Some of Our Contributions:

Berkeley UnixThe first free Unix, virtual memory, foundation of Linux

Computational complexityNP-completeness

CryptographyFoundations of cryptographic protocols

Approximation hardnessPCP (Probabilistically Checkable Proofs)

DevicesFINFET transistor, organic semiconductors, etc.

Electrical ground fault interruptorsInvented at Berkeley in the 1950s

Electronic design automation (EDA)Berkeley built this industry

Embedded systemsConcurrency, real-time computing, formal foundations

Floating pointIEEE 754 floating point standard

Graph algorithmsNetwork Flow, Planar separators and embeddings

Hybrid systemsMixed discrete/continuous systems

Nanoscale electronicsPhotolithography, transistors, transistor models, etc.

NetworkingTCP/IP, foundation of the Internet, in Berkeley UNIX

Mixed-signal circuitsKey contributions that make CMOS dominant

• MEMS systemsmicroelectromechanical systems

• Model-based designConcurrent models of computation, formal foundations

• Modern probabilistic AIReunified AI, learning, vision, control theory, stats

• Open source movementBerkeley software is truly free (vs. MIT’s GPL)

• Quantum computingFoundations of Quantum Complexity Theory

• Parallel computingNetwork of Workstations

• RAID storage systemsDominant design for large storage systems

• Randomized algorithmsRandomness as a computational resource

• Relational databasesAn EE/CS collaboration (Stonebraker & Wong)

• RISC processorsReduced instruction set computers

• Sensor NetworksBerkeley created this field

• Soft computingFuzzy logic

• Systems theoryFoundations of control, communications, signal proc.

• SpiceWorldwide standard in circuit simulation

60

Some of Our Contributions:

Berkeley UnixThe first free Unix, virtual memory, foundation of Linux

Computational complexityNP-completeness

CryptographyFoundations of cryptographic protocols

Approximation hardnessPCP (Probabilistically Checkable Proofs)

DevicesFINFET transistor, organic semiconductors, etc.

Electrical ground fault interruptorsInvented at Berkeley in the 1950s

Electronic design automation (EDA)Berkeley built this industry

Embedded systemsConcurrency, real-time computing, formal foundations

Floating pointIEEE 754 floating point standard

Graph algorithmsNetwork Flow, Planar separators and embeddings

Hybrid systemsMixed discrete/continuous systems

Nanoscale electronicsPhotolithography, transistors, transistor models, etc.

NetworkingTCP/IP, foundation of the Internet, in Berkeley UNIX

Mixed-signal circuitsKey contributions that make CMOS dominant

• MEMS systemsmicroelectromechanical systems

• Model-based designConcurrent models of computation, formal foundations

• Modern probabilistic AIReunified AI, learning, vision, control theory, stats

• Open source movementBerkeley software is truly free (vs. MIT’s GPL)

• Quantum computingFoundations of Quantum Complexity Theory

• Parallel computingNetwork of Workstations,

• RAID storage systemsDominant design for large storage systems

• Randomized algorithmsRandomness as a computational resource

• Relational databasesAn EE/CS collaboration (Stonebraker & Wong)

• RISC processorsReduced instruction set computers

• Sensor NetworksBerkeley created this field

• Soft computingFuzzy logic

• Systems theoryFoundations of control, communications, signal proc.

• SpiceWorldwide standard in circuit simulation

61