Last major update: June 28, 2005
Link to the K-1 Product Summary added on April 15, 2023
starting more updates in September 2023
... under construction ...
corrections are welcome
"Curt and Tom were considered off-the-wall crazy because it was well known that the big computer companies would have done it if it had been possible."
-- Larry West, quoted in William Broad, Star Warriors
Thomas M. McWilliams
Founder, Chairman & CTO
Tom is a well-known and well-respected computer architect with over 25 years in the computer industry. Prior to founding PathScale, from 1996 to 2001, Tom was a Distinguished Engineer and Principal Investigator at Sun Microsystems working on Server Architecture and advanced CAD tools. From 1993 to 1996, he was a Director in the MIPS division of Silicon Graphics managing microprocessor development and the MIPS architecture group. From 1989 to 1992, he was a Vice President at Amdahl in the Systems Architecture group. Before Amdahl, he founded Key Computer Laboratories in 1987. Amdahl acquired Key Computer, a company that designed of one of the first superscalar pipelined architecture computers, in 1989. In 1981 Tom co-founded Valid Logic Systems, whose products were based around his Ph.D. thesis work at Stanford University. He won the 1984 IEEE "W. Wallace McDowell Award" for the creation of the SCALD design methodology, which is the basis for most of the successful computer-aided engineering systems used in the industry. Prior to Valid, he was at the Lawrence Livermore National Laboratory (LLNL), where he was one of the principal designers of the S-1 supercomputer, a joint effort between LLNL and Stanford University. Dr. McWilliams received his Ph.D. in Computer Science from Stanford. He received his B.S.E.E and M.S.E.E. in Electrical Engineering from Carnegie Mellon University.
Tom's entry in the Hertz Fellows
Tom's entry in the LLNL Entrepreneurs' Hall of Fame
L. Curtis Widdoes, Jr.
Chairman & CTO
0-In Design Automation
Dr. Widdoes is widely recognized as a pioneer in the computer-aided engineering industry. Prior to 0-In, he founded two successful EDA companies, Logic Modeling Systems Inc. (1987) and Valid Logic Systems Inc. (1981). Dr. Widdoes holds a B.S. in engineering and applied science from the California Institute of Technology and a Ph.D. in computer science from Stanford University. He holds ten patents and is an IEEE W. Wallace McDowell Award recipient.
"Lowell was a real inspiration and a great mentor to all of us."
-- Curt Widdoes
From www.llnl.gov/IPandC/about/reports/ 2001_LLNL_IP_Report.pdf
In 1981, LLNL scientists Curt Widdoes and Tom McWilliams founded Valid Logic Systems, which, along with Daisy Systems and Mentor Graphics, is often credited with pioneering the electronic design automation (EDA) industry. Valid Logic Systems' initial products were based on the Structured Computer-Aided Logic Design (SCALD) software, developed at LLNL in the mid-70s. SCALD allowed almost completely automated design, manufacturing and testing of high performance digital circuits, drastically reducing the time and cost to design computers and digital circuits. Valid Logic reached a market capitalization of $250 million, considered a huge value for a start-up company in the 1980s, and was later bought by Cadence Design Systems. The significance of the SCALD technology was acknowledged by the respected professional organization IEEE Computer Society, which, in 1984, granted Widdoes and McWilliams the coveted W. Wallace McDowell Award, the second time this award was ever given for government-sponsored research.
"From the Valley of Heart's Delight to the Silicon Valley: A Study of Stanford University's Role in the Transformation," Carolyn Tajnai, December, 1996
VALID LOGIC SYSTEMS
Curt Widdoes (CS PhD'81), Tom McWilliams (CS PhD'80), Jeff Rubin, Jerry Anderson and Ray King founded Valid Logic Systems in 1981, and shortly afterward recruited Lou Scheffer. Their goal was to create tools for engineers to use in designing electronic systems, the beginning of the CAE (Computer Aided Engineering) industry. While at Stanford, Widdoes and McWilliams worked on the S-1 project under the direction of Professor Forest Baskett. According to Dr. Widdoes, "The S-1 Project that we worked on with Forest was a project to design and build multiprocessor supercomputers at Lawrence Livermore National Laboratory. The project team built three generations of ECL supercomputers from 1975 through 1985: the S-1 Mark I, Mark IIA, and Mark IIB. A single S-1 Mark IIA CPU contained nearly 30,000 ECL-100K integrated circuits, and was designed to be connected into clusters of 16 processing elements. The project also produced the SCALD computer-aided design software, which formed the basis of Valid Logic Systems in 1981." In 1984, Widdoes and McWilliams received the IEEE McDowell Award for the development of SCALD which formed the basis of the CAE industry. Valid, Daisy, and Mentor Graphics were the first major players in the CAE industry. Daisy, which later folded, was the first to go public; Valid went public in 1983 and Mentor Graphics followed soon after. In 1987, Curt Widdoes left Valid, which was bought by Cadence in 1989.
Logic Modeling Systems
In 1987, Curt Widdoes and Steve White co-founded Logic Modeling Systems. While at Valid, Widdoes had invented hardware modeling technology and received fundamental patents covering it. He bought exclusive rights to the patents from Valid and founded Logic Modeling to bring a product to market. Logic Modeling achieved a monopoly in hardware modeling. Then in 1992, Logic Modeling Systems merged with Logic Automation, which sold software models of components. The new company was called Logic Modeling Corporation.
Logic Modeling Corporation
Widdoes was named President and later Chairman of the new company, which dominated the system level components modeling market. In 1994, Widdoes sold Logic Modeling Corporation to Synopsys for approximately $120 million, and then left Synopsys in July 1994 to pursue other interests.
0-In Design Automation
In June, 1996, Widdoes, White, David Dill, (CS Professor), Richard Ho (CS PhD'96), and Paul Estrada founded 0-In Design Automation to provide better tools for the functional validation of electronic designs, which has become the primary bottleneck in the development of large integrated circuits. Products from 0-In should be available in 1997.
"A Policy for Government Support of Computer Systems R & D: A Look at 50 Federally Funded Computer Systems Research Projects Over 30 Years," Gordon Bell, April 1994
Lowell Wood's group at Livermore worked on the development of several
multiprocessor computers, two models of its S-1 machine were built,
but the only real output was a method of interconnecting chips on a
substrate resulting in the startup company NCHIP.
Stanford & Lawrence Livermore S-1 & Logic Timing Verifier. Two PhD students, Tom McWilliams and Curt Widdoes worked on the design of a large scale multiprocessor project. The principle technical output of the design was the first Timing Verifier, Scald. In the early 1980s, this became a key component of a design system that launched Valid Logic, one of the first three startup CAE companies. The S-1 became a project at Livermore, and a prototype was built that validated the design system under Lowell Wood. A more ambitious full-scale multiprocessor was attempted and one processor was built.
Curt's entry in the Hertz Fellows
Curt's entry in the LLNL Entrepreneurs' Hall of Fame
see LLNL tribute to Haussman
special issue of Science and Technology Review, January-February 1999, https://str.llnl.gov/content/pages/past-issues-pdfs/1999.01.pdf
Advisees: Mike Farmwald, Tom McWilliams, and Curt Widdoes
Advisees: Fernando Castaneda, Fred Chow, Eric Gilbert, Arthur Keller, Peter Nye, Armando Rodriguez, and Dan Sleator
Advisees: Brent Hailpern and Bruce Hitson
Note that Jeff Barth and John McCarthy were Stanford faculty who were also listed as Principal Investigators on S-1-related grants to Stanford.
"As for the basics, I started at the S-1 in Feb 1979, right after getting a BS from MIT. I left in June of 1988, and took a staff position at CMU to work on the AFS project. That spun-off to become Transarc, which was then acquired by IBM, where I still am. I came to S-1 to work on the Amber operating system, but did lots of other things as well including working on CAD software and debugging hardware."
Rodney A. Brooks
Rodney A. Brooks is Director of the MIT Computer Science and Artificial Intelligence Laboratory, and is the Fujitsu Professor of Computer Science. He is also Chairman and Chief Technical Officer of iRobot Corp.
Jeffrey M. Broughton
Founder, Vice-President of Engineering
An innovative and experienced software engineering manager and computer architect, Jeff has spent over 25 years focused on high end computing. Most recently, he led the software effort in Sun Microsystems Laboratory to develop a massively parallel simulation engine for in-house use. Prior to Sun, he was at Amdahl Corporation, serving as Manager of Architecture and alternative platforms based on SPARC and Intel architectures. He worked broadly within the worldwide Fujitsu group of companies, and contributed to the development of the SPARC Version 9 architecture. Amdahl had acquired Key Computer, where Jeff was responsible for development of an optimizing compiler suite for a superscaler supercomputer. Earlier, Jeff served as Technical Director of the S-1 Supercomputer Project at Lawrence Livermore National Laboratory, and before that was Project Engineer in charge of compilers and operating system software. He began his career as an engineer in the Multics group at Honeywell Information Systems. Jeff is an inventor of optimistic concurrency protocols, distributed cache coherence protocols, domain partitioning mechanisms, and software methods for cycle-based logic simulation. Jeff received concurrent S.B. and S.M. degrees in Electrical Engineering and Computer Science from M.I.T.
Inteview of Jeff by Dan Olds, "Hyperion Jeff Broughton Interview," Oct. 27, 2020.
John D. Bruner
Dr. Bruner worked on the S-1 project from February 1983 through August of 1988. He is now at Motorola.
Hon Wah Chin
Dr. Chow has over 24 years of experience developing advanced optimization techniques in production compilers. Recently, he lead a team to develop code generation techniques for Cognigine's Variable Instruction Set Architecture, which was successfully deployed in the compiler for Cognigine's network processor chip. Earlier, Fred was Chief Scientist at SGI, where he spearheaded SGI's compiler efforts to achieve highest performances in SGI's systems. He was Chief Architect of the Pro64 Compiler, which was re-targeted to the Itanium and made available to the public as the Open64 Compiler. Before SGI, Fred was the original member of the compiler team at MIPS Computer Systems. MIPS' Ucode Compiler was a productization of Fred's pioneering Ph.D. research at Stanford. MIPS' Ucode Compiler became the very first commercial compiler that incorporates a full set of global optimization capabilities, and established highly optimizing compiler as an indispensable component in modern microprocessors. Fred received his B.Sc. degree from the University of Toronto in 1977. He received his M.S. and Ph.D. degrees from Stanford University in 1978 and 1983, respectively. Throughout his career, Fred has put major emphases on compiler infrastructures that enable efficient and stable optimization implementation. Fred has written numerous articles in journals and research publications. He holds 8 patents.
-- bio from PACT 2002 Tutorial
The PathScale compiler development team is led by Dr. Fred Chow, formerly chief scientist for compilers at SGI, and recognized as one of the world's leading authorities on compiler technology.
-- Pathscale media release
[packaging and cooling for the AAP]
Dr. Farmwald is highly respected as an entrepreneur with one of the most successful track records in Silicon Valley. Known for his unique combination of computer engineering skill and market vision, Dr. Farmwald has founded six companies to date, holds fifty U.S. patents, and has approximately twenty patents pending. Early in his career, with a B.S. degree in mathematics from Purdue University and a Ph.D. in computer science from Stanford University, Dr. Farmwald joined Lawrence Livermore National Laboratory as an architect on the S1 supercomputer. In 1986, he co-founded FTL, a supercomputer company that merged with MIPS Computer Systems that same year and served as Chief Scientist for High End Systems. After MIPS, Dr. Farmwald was named an Associate Professor of Electrical and Computer Engineering at the University of Illinois and began work on Rambus, a company he co-founded to address the performance gap between microprocessors and the memory they rely on to obtain data. At Rambus, Dr. Farmwald served as Vice President and Chief Scientist. Other companies Dr. Farmwald has founded include Chromatic Research, a multimedia accelerator company acquired by ATI, and Epigram, the home networking company acquired by Broadcom. In 1998, Dr. Farmwald co-founded Matrix Semiconductor. Dr. Farmwald currently sits on the boards of Rambus (Nasdaq: RMBS), Matrix Semiconductor, and AON Networks, is a Benchmark Capital Venture Partner, and is a General Partner at Skymoon Ventures.
Mike's entry in the Hertz Fellows
Mike's entry in the LLNL Entrepreneurs' Hall of Fame
Dr. Richard Gabriel is a leader in the Lisp/OOP community with years of contributions to standardization and founding a highly successful company, Lucid.
-- ACM Fellows citation
Gary W. Hagensen
Associate Director, Computer Science
I'm Brent Hailpern, a Research Staff Member and Associate Director, Computer Science at the IBM Thomas J. Watson Research Center, in Hawthorne, New York. I have worked at IBM since 1980, with most of that time at IBM Research. My personal research interests are in programming languages, concurrent systems and object-oriented systems. I have managed projects and departments in programming languages, software engineering and environments, operating systems, multimedia systems, Internet and pervasive technologies, workflow, email, and K-12 education. I have also worked in IBM corporate headquarters and as a software product manager. I am currently an Associate Editor for ACM's Transaction on Programming Languages and Systems (TOPLAS) and a Fellow of the ACM and the IEEE. I am a past Chair of ACM SIGPLAN, a past Steering Committee Chair of the OOPSLA conference, and a past Secretary of the ACM.
Dr. Hitson is an experienced entrepreneur and businessman. He has founded or helped grow successful technology-intensive companies in Internet e-commerce, Internet derivatives trading, computer integrated telephony, call center applications, and statistical data mining/warehousing. Hitson's expertise and passions are defining and managing new business and technology initiatives that allow order-of-magnitude changes in the way businesses operate and profit. As chief of technology for Teknekron Infoswitch, he started a computer telephony integration business unit that achieved multi-million dollar sales in its first year of operations. At Integral, he Internet technology to the financial derivatives industry, a key strategic component in out both a technical and marketing niche for an upcoming IPO. At EIT he created the initial for electronic commerce that took them from an R&D start-up to a $28 million acquisition. Hitson received his doctoral degree from Stanford University based on a dissertation "Knowledge-Based Monitoring and Control of Distributed Systems". As part of this work he conceived, built, and analyzed a practical expert system for diagnosing protocol-level problems in a distributed collection of TCP/IP networks. He also holds several patents in areas related to computer integrated telephony.
Eric H. Jensen
Chief Executive Officer
Eric Jensen co-founded ScoreBoard, Inc., in October 1993. He is responsible for corporate funding, daily operations, and corporate strategy. Mr. Jensen has over 15 years of experience in management, software development, and R&D.
Prior to founding ScoreBoard, Inc., Mr. Jensen served as an original executive at PSINet (NASDAQ: PSIX), the world's first Internet Service Provider. While at PSINet, Mr. Jensen created PSILink, the first Internet access service for beginning Internet users. This service offered Internet mail, news and file transfer services, and supported wireless Internet access over a national wireless data network deployed by RAM Mobile Data. Mr. Jensen also managed the development of the first complete service management tool focused on supporting Internet Protocol (IP) over a frame relay based network.
Prior to joining PSINet, Mr. Jensen was a project leader at Sun Microsystems (NASDAQ: SUNW) where he developed and received a patent for the super computing architecture design which became a key contribution to the development of SUN's SPARC multiprocessor memory model.
Prior to joining SUN, Mr. Jensen served as a project leader at the Lawrence Livermore National Laboratories. While there, Mr. Jensen led the development of super computing capabilities based on multi-processor based architectures.
Mr. Jensen holds 10 US Patents, including US Patent No. 5,265,233 - Method and Apparatus For Providing Total and Partial Store Ordering For a Memory In Multi-Processor System and US Patent No. 5,219,614 - Real time, Concurrent, Multifunction Digital Signal Processor Subsystem For Personal Computers.
Mr. Jensen earned a Bachelor of Science in Electrical Engineering in 1983 as well as a Master of Engineering in 1984 from Rensselaer Polytechnic Institute (RPI).
Earl Killian's early work was in the software industry on networking, compilers, operating systems, and binary translation. In the last sixteen years, he has put his system software experience to work in computer architecture, designing instruction-set architectures, pipelines and performance models for microprocessors. As MIPS's Director of Architecture, he designed the MIPS III 64-bit instruction-set extension, and led the work on the R4000 microarchitecture. He was a cofounder of QED, which created the R4600 and R5000 MIPS processors. Most recently he was chief architect at Tensilica working on configurable/extensible processors.
Mr. Killian is currently independent, pursuing alternative energy, construction, and transportation interests. He also serves as technical advisor to several start-ups and does volunteer work (most recently teaching at Stanford).
Mr. Killian holds a B.S.E.E from MIT.
"Finally, on a personal note, the oversized nature of the S-1 led several of us to work on the S-2 (though we couldn't call it that politically), which was 2 boards of ECL instead of 72. It was essentially a RISC architecture, though we did not know that acronym at the time. This led me to add architecture statistics gathering tools to Pastel to guide the design, especially basic-block counting and cache trace generation. Later at MIPS I revived these ideas at the binary translation level when I created moxie, the MIPS to VAX translater/instrumenter, and pixie, the MIPS to MIPS instrumenter. These became the basis for all the later MIPS R3000 and R4000 performance simulations."
-- Earl Killian
As General Manager, John Manferdelli is responsible for providing the strategic direction for the development of Microsoft's Trusted Platform Technologies. Before that, he was a Senior Researcher and Architect at Microsoft. His professional interests are cryptography, operating systems, emerging technologies, combinatorial mathematics, digital rights management and other security and privacy related technology. Prior to Microsoft, John was the a founder, and at various times Vice President of R&D and CEO of Natural Language Incorporated, which was eventually acquired by Microsoft. John has also held many senior level technology positions at TRW, Lawrence Livermore National Labs and Bell Labs and taught at Stevens Institute of Technology. John has his BS in Physics from Cooper Union for the Advancement of Arts and Sciences, and a PhD from the University of California at Berkeley.
-- bio from Microsoft Security Workshop 2002
[Tom's younger brother; worked on wafer-scale integration]
Tessera also today announced the appointment of a new president and CEO,
Bruce McWilliams, formerly chairman of S-Vision Inc., a company that
pioneered in silicon microdisplays.
McWilliams co-founded S-Vision in 1996 and has 18 years of technical experience in display and semiconductor technology, electronic packaging, and contract manufacturing. He was a senior vice president at Flextronics International, and was president and CEO of nCHIP Inc., which he co-founded in 1989. McWilliams also led a team at Lawrence Livermore National Laboratory that focused on electronic packaging, laser processing of semiconductors and optical systems for target tracking.
-- from Silicon Strategies, Sept. 1999
Bruce's entry in the LLNL Entrepreneurs' Hall of Fame
[worked on wafer-scale integration]
[wlater orked on the Exponential x704 design]
[later worked on the Exponential x704 design]
[Jeff] Rubin was on the S1 project back in 1978. He is a code-writing savant. I watched him write an emulator for the S1 computer in 4 days. The S1 was a complex instruction set, 16-processor machine with over 4000 instructions including matrix multiply and FFT, 32 non-regular registers, and dozens of addressing modes. He wrote the 60,000 line program in PDP-10 assembly language in 4 days. He debugged it for half a day and 30 people used it every day for 6 years and never found a bug. He wrote the optimizing assembler for it immediately after - 40,000 lines in 3 days, a few hours of debugging and no bugs found in 6 years.
Jeff's entry in the LLNL Entrepreneurs' Hall of Fame
Key Computer Labs
Key Computer Labs was a supercomputer start-up in Silicon Valley
in the late 1980s. The founders (Jeff Rubin and Tom McWilliams)
were veterans of the Livermore Labs' S1 project, who had gone on
to make a bit of money in commercializing the design tools (SCALD,
if I remember correctly) that had originally been developed for
the S1. Key raised enough capital to get going, but soon found
themselves short enough of cash that they were rescued by Amdahl.
It has been speculated that this was an insurance policy against
IBM delivering new technology into the mainframe marketplace
sooner that Amdahl expected. Amdahl ultimately stopped the
supercomputer project and turned Key into something of an
advanced architecture lab.
The Key machine was to have been a heavily pipelined design, with a form of predicated execution as a mechanism to avoid branch bubbles.
2023 update: the K-1 Product Summary from 1988 has been added to the Bitsavers collection
2/15/79 - 8/1/80 Lawrence Livermore National Labs
Member of O Group, run by Lowell Wood. Designed, developed, and began to implement paging, virtual memory, segmentation, and file system for the Amber timesharing system for the S-1 Mark II supercomputer then being developed at Livermore.
[worked on wafer-scale integration]
John has a MSEE from Stanford. He started as a digital design engineer and then changed to software. His last "real" fulltime job was at LLNL, building desktop supercomputers with the same folks that ultimately brought us Brilliant Pebbles, Rambus, SCALD and Valid Logic, and ONI, among other Valley successes. He has been a consultant / contractor since 1983.
-- ACCU speaker bio
[S-1 page] [Mark's homepage] [CPSC homepage] [Clemson Univ. homepage]