Neil J. Gunther

From Wikipedia, the free encyclopedia

Neil James Gunther
Neil Gunther at Bletchley Park 2002 "A quantum leap is neither"
Neil Gunther at Bletchley Park 2002
"A quantum leap is neither"
Born August 15, 1950 (1950-08-15) (age 57)
Preston, Victoria, Australia
Residence Flag of the United States United States, California,
Nationality Flag of Australia Australia
Fields Computational information systems (classical and quantum)
Institutions San Jose State University
Syncal Corporation
Xerox Palo Alto Research Center
Performance Dynamics Company (Founder)
Ecole Polytechnique Federale de Lausanne (EPFL)
Alma mater La Trobe University
University of Southampton
Doctoral advisor Tomas M. Kalotas (Honors)
Christie J. Eliezer (Masters)
David J. Wallace (Doctorate)
Known for Performance analysis
Capacity planning tools
Theory of large transients
Universal scalability law

Neil Gunther, (born August 15, 1950) is a computer information systems researcher best known internationally for developing the open-source (freeware) performance modeling software called PDQ (Pretty Damn Quick), before open-source was trendy. He has also been cited for his contributions to the theory of large transients in computer systems and packet networks, and his universal law of computational scalability. He is currently focused on developing quantum information system technologies.

Contents

[edit] Biography

Neil James Gunther, M.Sc. (LaT), Ph.D. (Soton) was born in Melbourne, Australia. He holds undergraduate degrees in Chemistry and Physics, a Masters Degree in Applied Mathematics (1976) from La Trobe University, Australia, and a doctorate in Theoretical Physics (1980) from the University of Southampton, England. Since he was a research student of Prof. C. J. Eliezer (one of Dirac's few research students), he can claim a Dirac number of 2.

Gunther is a member of the Association for Computing Machinery (ACM), American Mathematical Society (AMS), American Physical Society (APS), Computer Measurement Group (CMG), Institute of Electrical and Electronics Engineers (IEEE), Institute for Operations Research and Management Sciences (INFORMS), and SIGMETRICS.

[edit] German and Scots Ancestry

Gunther's father, Walter August Gunther (born 1908 in Clifton Hill, Victoria), is third generation Australian of German descent. The original immigrants were the family members of Friedrich Christoph Gunther (born 1807 in Glendelin, Demmin, Prussia), who came to Australia, probably between 1848 and 1853, from the town of Demmin, Pommern (Pomerania) in the area of north eastern Germany then known as Prussia and now part of Mecklenburg-Vorpommern. They appear to have settled in the gold fields of central Victoria at a place called "McCallum's" (probably an abbreviation for McCallum's Gold Fields Common[1]) along McCallum's Creek (now the town of Craigie), near the city of Maryborough. Friedrich's son, August Philipp Gunther (born 1832 in Prussia) married Elise Auguste Bruhn (born 1836 in Dresden, Germany) in central Victoria in 1855 and thus, the Gunthers and the Bruhns became related by marriage. Dr. George H. Bruhn, a German physician and the father of Elise Bruhn[2], was among the first to discover gold in Victoria. His discovery[3] near the town of Clunes was recorded in 1851. The Victorian "gold rush" peaked around 1854, some five years after the gold rush in California.

Gunther's mother, Margaret Wood MacFadyen (Gaelic: Macphaidein, meaning Son of Paidin, a diminutive form of Padraig (Patrick)), came to Melbourne, Australia at the age of 15 when her family emigrated from Glasgow, Scotland in 1929. The MacFadyen's were the first Sept of the Clan Mclaine of Lochbuie from the Isle of Mull on the west coast of Scotland. The name MacFadyen was first recorded in Kintyre in 1304. The McFadyens were the original owners of the lands at Lochbuie and were also known as Sliocht Nan Or-Cheard (The Tribe of Goldsmiths) on the Isle of Mull. The current and 26th hereditary clan Chieftain is Lorne Maclaine.

[edit] School Years

Gunther attended both Preston East Primary School (1955-1956) and Balwyn North Primary School (1956-1962). When he began to learn his "sums" (arithmetic addition) by rote, Gunther was struck by the certainty with which statements like 3 + 4 = 7 could be made. If you are presented with the numbers 3 and 4 and asked for the sum, the result is always 7. This seemed to stand out in complete contradiction to everything else he knew as a child at the time. The weather, for example, could be sunny while walking to school but overcast or wet while walking home (especially in Melbourne). At some point it occurred to him that if he was presented with the single number, say 7, he could not tell which two or more numbers might have been summed to produce it. This uncertainty was more consistent with everything else he knew about the world, but was never presented in school; remaining hidden behind the relentless rote learning of sums and rules. Moreover, he was too young to formulate his observation in the way it is presented here. Consequently, this notion of a hidden uncertainty lurking behind "official" arithmetic remained as a kind of nagging doubt, which led to a kind of mistrust of rules that come from authoritarian sources. Only many years later would he find out about algebra (which seemed more transparent because symbols could not be coalesced so easily) as well as the theory of partitions[4] and their relationship to probability, randomness and models of uncertainty; the very things that have entered into his current research activities.

For his tenth birthday, Gunther received a copy of the now famous book entitled The Golden Book of Chemistry Experiments from an older cousin. Thoroughly inspired by this marvelous book, he started working through these experiments by making use of various chemicals that could be found around the house. Recognizing the success of the birthday present, the brother of the same cousin presented Gunther with a chemistry set the following Christmas. All went well until he accidentally spilt some potassium permanganate solution on his bedroom carpet. Thereafter, his mother banished him to an alcove in the garage. With this new seclusion, things really took off to the point where he created a small laboratory, replete with industrial chemicals (including plenty of carcinogens and poisons that would not be possible for anyone to purchase over-the-counter today, let alone a child) and second-hand glassware often purchased from H.B. Selby's Chemical Supplies on Swanston Street in Melbourne. Unlike most of his peers who only had a passing interest in chemistry, Gunther was not interested in making things that exploded but rather, finding out how things like detergents and oils were composed by "cracking" them in his fractionating column.

Gunther attended Balwyn High School and it was in junior art classes that he gained a reputation for his special blue tone which he developed using some now forgotten mixture of titanium white, cobalt blue and prussian blue. In that same period, a new First Form (7th grade) science teacher wrote down the Alkane series on the blackboard. Riveted by the pattern that emerged, Gunther realized he had not seen anything like this in his precious "Golden book". In fact, the whole subject of organic chemistry was only alluded to under the chapter on Carbon. His father, being the Superintendent of the Melbourne City Council electrical power station, managed to get a loan of an organic chemistry text[5] from the chemists in the quality control laboratory. This ultimately led to an intense interest in synthesizing Azo dyes[6], of which paranitraniline red was his favorite. The crescendo of all this chemical activity, at around age 14, was an attempt to predict the color of azo dyes based on the chromophore-auxochrome combination. Apart from drawing up empirical tables, this effort was largely doomed because Gunther had no knowledge of quantum theory; a position that was not helped by the fact that the textbook[5] he was relying on had originated in the 1930s, well before the concept of molecular orbital theory began to appear in chemistry books. Only very recently, since becoming involved with quantum imaging and photonics, has he realized how close all this was to modern color theory and the trail blazed by such eminent figures as Newton, Young, Goethe and Schrödinger.

[edit] Post-Doc Years

Gunther taught physics at San Jose State University from 1980-1981. In 1981, he joined Syncal Corporation, a small company contracted by NASA and JPL to develop thermoelectric materials for their deep-space missions. Because of the huge distances involved, NASA's deep-space missions cannot utilize solar radiation to produce power. Instead, they must carry their own on-board power generation systems. In the case of the Voyager 1, Voyager 2 and Galileo missions, the electrical power is provided by a radioisotope thermoelectric generator (RTG) which utilize the Seebeck effect to produce electric current via the thermal gradient. On the Voyager 1 spacecraft, the RTGs appear as three canisters on one of the boom arms.

Gunther was asked to analyze the thermal stability test data from the Voyager RTGs. He discovered that the stability of the silicon-germanium (Si-Ge) thermoelectric alloy was controlled by a soliton-based precipitation mechanism[7]. Over a very long period of time, the thermal gradient eventually drives the germanium out of solution, thereby reducing the effective power output. Even with diminished power output, the RTG's still enable command and control communications 30 years after launch!. This work was important because JPL wanted to use it as a foundation for selecting the next generation of RTG materials for the Galileo mission, which was started in 1981, shelved, resurrected, and finally launched in 1989. Syncal was bought out by Thermo Electron Corporation in 1982.

[edit] Xerox Years

In 1982, Gunther joined Xerox PARC (no longer owned by Xerox Corporation) to develop parametric and functional test software for PARC's small-scale VLSI design fabrication line. Ultimately, he was recruited onto the Dragon multiprocessor workstation project where he also developed the PARCbench multiprocessor benchmark. This was his first fore into computer performance analysis. Starting in 1987, he developed a Wick-rotated version of Feynman's quantum path-integral formalism[8] for analyzing performance degradation in large-scale computer systems and packet networks. The key observation is that the transient stability of the queueing system strongly resembles a quantum tunneling process.

Others have investigated the same computer performance problems using alternative mathematical techniques such as Large deviations theory[9] and Catastrophe Theory[10]. The same concepts have also been applied to understanding rare events in financial markets[11] e.g., the catastrophic failure of Long Term Capital Management.

[edit] Pyramid Years

In 1990 Gunther joined Pyramid Technology (now part of Fujitsu Siemens Computers) where he held positions as Senior Scientist and Manager of the Performance Analysis Group that was responsible for attaining industry-high TPC benchmarks on their Unix multiprocessors. He also performed simulations for the design of the Reliant RM1000 parallel database server.

[edit] Consulting Practice

Gunther founded Performance Dynamics Company as a sole proprietorship, registered in California in 1994, to provide consulting and educational services for the management of high performance computer systems with an emphasis on performance analysis and enterprise-wide capacity planning. He went on to release and develop his own open-source performance modeling software called "PDQ (Pretty Damn Quick)" starting around 1998. That software also accompanied his first textbook on performance analysis entitled The Practical Performance Analyst. Several other books have followed since then.

[edit] Current Research Interests

[edit] Quantum Information Systems

Starting in 2004, Gunther has embarked on joint research into quantum information systems based on photonics. During the course of his research in this area, he has developed a theory of photon bifurcation[12] that is currently being tested experimentally at École Polytechnique Fédérale de Lausanne. This represents yet another useful application of Feynman's quantum path integral to conveniently circumvent the old chestnut concerning the wave-particle duality of light. In its simplest rendition, this theory can be considered as providing the quantum corrections to the Abbe-Rayleigh diffraction theory[13] of imaging and the Fourier theory of optical information processing.

[edit] Performance Visualization

Inspired more by the work of Tukey than Tufte, Gunther has explored ways to help the systems analyst visualize performance in a manner similar to that already available in scientific visualization and information visualization. One such tool he developed, called Barry[14], employs barycentric coordinates to visualize sampled CPU utilization data on large-scale multiprocessor systems. More recently, he has applied the same 2-simplex barycentric coordinates to visualizing the Apdex application performance metric, which is based on categorical response time data. A barycentric 3-simplex (a tetrahedron), that can be swiveled on the computer screen using a mouse, has been found useful for visualizing packet network performance data.

[edit] Universal Law of Computational Scalability

The relative capacity C(N) of a computational platform is given by:

C(N) = \frac{N}{1 + \alpha (N-1) + \beta N (N-1)}

where N represents either:

  1. the number of real users, virtual users or load generators driving a fixed hardware configuration. In this case, the number of users acts as the independent variable while the processor configuration remains constant over the range of user load measurements. This usage is appropriate for predicting software scalability.
  2. the number of physical processors or computational nodes in the hardware configuration. In this case, the number of user processes executing per processor is assumed to be fixed across every added processor. For example, the saturation value was 10 processes per processor, then on a 4 processor platform you would load the system with 40 virtual users. This usage is appropriate for predicting hardware scalability.

The parameters α and β represent respectively the levels of contention and coherency-delay in the system taken as a whole. The β parameters also quantifies the retrograde throughput seen in many stress tests but not accounted for in either Amdahl's law or event-based simulations. This scalability law was originally developed by Gunther in 1993[15] while he was employed at Pyramid Technology.

The three terms in the denominator of C(N) correspond respectively to concurrency (degree of parallelism), contention (queueing for shared resources), and coherency (thrashing effects). Since there are no topological dependencies, C(N) can model symmetric multiprocessors, multicores, clusters, and GRID architectures. Also, because each of the three terms has a definite physical meaning, they can be employed as a heuristic to determine where to make performance improvements in hardware platforms or software applications.

The universal scalability law can be derived from conventional queueing theory using the following theorem:

Theorem (Gunther 2002): Amdahl's law for parallel speedup is equivalent to the synchronous queueing bound on throughput in the Repairman Model[16] of a multiprocessor.

The proof was first given in A New Interpretation of Amdahl's Law and Geometric Scalability.

Conjecture: Gunther's equation for C(N) is a rational function with parameters α and β. Two parameters are necessary and sufficient for any computational scalability model based on rational functions.

[edit] Computational Mathematics

Over the past fifteen years, Gunther has had an abiding interest in of the 3x+1 problem, not with the goal of developing a technical proof of the original conjecture but rather, using computers as a tool to examine it for structure that might lead to better computer-generated visualizations of this and related problems in number theory. In one early attempt along these lines he employed VRML[17]. Paul Erdös famously stated about the 3x+1 problem, "Mathematics is not yet ready for such problems." Gunther thinks that perhaps computers are.

More formally, Gunther has developed a functional Diophantine equation that generalizes Terra’s theorem (1976) and is based on a graphical primitive: the G-set. The G-set is related to the predecessor sets of Wirsching [18] by the following theorem.

Theorem (Gunther 1999): The G-set (Gi) is a directed subgraph in ΓT (Collatz tree) formed by acyclic predecessor sets starting at b and terminating at vertex a with exactly k = 1 edges arising from T1(x) = (3x + 1) / 2, i.e.,

P^{(\#T_1=k)}_T(a) := {b \in P_T (a)}.

The proof is unpublished. This theorem leads to the following conjecture for the construction of ΓT.

Conjecture (Gunther 1999): \Gamma_T \equiv \cup_i G_i , where \forall i \in \mathbb{N}, enumerates all G-cells in ΓT such that the unique G-set G0 contains the degenerate cycle 2 \leftrightarrow 1.

Ironically, given his lack of intent to find a proof, the formal associations with the theorems of Terras and Wirsching, make it plausible that this method of sub-graph enumeration might form the basis of an inductive proof.

[edit] Musical Interests

Gunther was first taught to play banjo-mandolin by his father at the age of 11. Ultimately, it proved too difficult to capture the then emerging surfing sound of the 1960's on banjo, so Gunther bought his first very cheap guitar at age 13.

[edit] Guitar Styles

At age 15, Gunther formed his first rock band called Time's Up. At various points in time, some version of this band backed some well-known Australian pop-singers including: Ronnie Burns, Johnny Young and Olivia Newton-John. But Gunther's true love is modern jazz guitar, in general, and Brazilian samba and bossa nova, in particular. The latter he learnt from Helmut Becker in Melbourne, who also taught him classical guitar. It was here that Gunther developed a penchant for playing nylon-string, rather than steel-string, guitars with a classical-width fretboard. Later, he was taught jazz guitar based on George Russell's Lydian Chromatic technique[19]. He currently owns and plays:

Gunther has also written entries for the Blue Book of Guitars[20].

[edit] Cyber Band

Lacking time to participate properly in a real performing band, Gunther had always planned to settle more for home recording. Toward this end, he has a Power Macintosh G4 digital audio-workstation with twin disks. More recently, he has been working with musicians in Europe in the form of a cyber-band. Like email, immediacy of action is not required. Tracks are created and exchanged by individual musicians on the Internet, then arranged and mixed as time permits. The results can also be downloaded by any interested parties.

[edit] Awards

  • Summer Research Institute visitor, EPFL 2006 and 2007.
  • Nominated for the A. A. Michelson Award in 1997 and 2006.
  • Lecturer, Western Institute of Computer Science, Stanford University, 1997-2000.
  • Best paper award, CMG conference 1996.
  • Visiting Scholar in Materials Science, Stanford University, 1981-1982.
  • University of Southampton Advanced Studies Travel Grant, 1978.
  • Science Research Council Studentship, U.K. 1976-1980.
  • Commonwealth Postgraduate Scholarship, Australia 1975-1976.

[edit] Quotes

  • As a consultant, I offer more harangue for the buck.
  • Lately, I've been solicited to give so many talks I feel like Mister Ed The Talking Whore.
  • It's better to have wrong expectations, than no expectations.
  • Best Practices are an admission of failure.
  • A queue is a line of customers waiting to be severed.
  • The only dumb question is the one never asked.
  • A quantum leap is neither.
  • Art irritates life.[21]
  • If you want to be more productive, go to sleep.
  • All meaning has a pattern, but not all patterns have a meaning.

[edit] Selected Bibliography

[edit] Theses

  • The Feynman Path Integral in Non-Relativistic Quantum Mechanics and Quantum Electrodynamics, La Trobe University (AUS),

B.Sc. Honors dissertation, Department of Physics, Oct. (1974)

  • Dynamical Symmetry Groups: The Study and Interpretation of Certain Invariants as Group Generators in Quantum Mechanics, La Trobe

University (AUS), M.Sc. dissertation, Department of Applied Mathematics, Nov. (1976)

  • Broken Dynamical Symmetries in Quantum Field Theory and Phase Transition Phenomena, University of Southampton (U.K.), Ph.D.

dissertation, Department of Physics, Dec. (1979)

[edit] Books

  • The Practical Performance Analyst, McGraw-Hill, New York, New York 1998, ISBN 0079129463 (Out of print)
  • The Practical Performance Analyst, iUniverse.com Press, Lincoln, Nebraska 2000, ISBN 059512674X (Reprint edition)
  • Performance Engineering: State of the Art and Current Trends, Lecture Notes in Computer Science, Springer-Verlag

Heidelberg, Germany, October 2001, ISBN 3540421459 (Contributed chapter)

[edit] Invited Presentations

  • Goldstone Modes in First-order Phase Transitions, Sixth West Coast Conference on Statistical Mechanics, IBM Research Laboratories, San Jose, June (1980)
  • Instanton Techniques for Queueing Models of Large Computer Systems: Getting a Piece of the Action, SIAM Conference on Applied Probability in Science and Engineering, New Orleans, Louisiana, March (1990)
  • (Numerical) Investigations into Physical Power-law Models of Internet Traffic Using the Renormalization Group, IFORS Conference of Operations Research Societies, Honolulu, Hawaii, July 11-15 (2005)

[edit] Papers

[edit] References

  1. ^ National Library of Australia, Victoria. Mines Dept., Mining district of Maryborough, MAP RM 1730, 1861.
  2. ^ The late Emma Steers (Gunther's paternal grandmother), private communication.
  3. ^ David Blair (1879). "The History of Australasia," p. 413. McGready Thompson and Niven. 
  4. ^ J. Riordan (1967). "An Introduction to Combinatorial Analysis". John Wiley and Sons. ISBN 0486425363. 
  5. ^ a b F. Sherwood Taylor (1962). "Organic Chemistry" (5th edition). Heinemann.  (Originally published in 1933)
  6. ^ Thomas Maschmeyer. "Colors and Dyes".
  7. ^ Gunther, Neil J. (1982). "Solitons and Their Role in the Degradation of Modified Silicon-Germanium Alloys" in Proc. IEEE Fourth Int. Conf. on Thermoelectric Energy Conversion.. IEEE, Volume 82CH1763-2, Pages 89–95.
  8. ^ Gunther, Neil J. (1989). "Path Integral Methods for Computer Performance Analysis". Information Processing Letters, Volume 32(1) Pages 7–13.
  9. ^ A. Shwartz and A. Weiss (1995). "Large Deviations for Performance Analysis". Chapman and Hall. ISBN 0412063115. (Gunther's approach is cited on p. 28)
  10. ^ R. Nelson, "Stochastic Catastrophe Theory in Computer Performance Modeling," J. ACM, Volume 34, Page 66 (1987)
  11. ^ N. N. Taleb (2007). "The Black Swan: The Impact of the Highly Improbable". Random House. ISBN 978-1400063512. (Large deviations theory presented for the non-mathematician)
  12. ^ Gunther, Neil J. , E. Charbon, D. L. Boiko, and G. Beretta (2006). "Photonic Information Processing Needs Quantum Design Rules". SPIE Online.
  13. ^ E. G. Steward (2004). "Fourier Optics: An Introduction". Dover. ISBN 0-486-43504-0. 
  14. ^ Gunther, Neil J. (1992). "On the Application of Barycentric Coordinates to the Prompt and Visually Efficient Display of Multiprocessor Performance Data" in Proc. VI International Conf. on Modelling Techniques and Tools for Computer Performance Evaluation, Edinburgh, Scotland. Antony Rowe Ltd., Wiltshire, U.K., Pages 67–80. ISBN 0-7486-0425-1. 
  15. ^ Gunther, Neil J. (1993). "A Simple Capacity Model for Massively Parallel Transaction Systems" in Proc. CMG Conf., San Diego, California. CMG, Pages 1035–1044.
  16. ^ D. Gross and C. M. Harris (1998). "Fundamentals of Queueing Theory". Wiley-Interscience. ISBN 0-471-17083-6. 
  17. ^ Gunther, Neil J. (2000). "Seeing the Forest in the Tree: Applying VRML to Mathematical Problems in Number Theory" in Proc. IEEE-SPIE 12th International Symposium on Internet Imaging.. SPIE International Society for Optical Engineering, Volume 3964.
  18. ^ Wirsching, Günther J. (1998). "The Dynamical System Generated by the 3n+1 Function". Springer Lecture Notes in Mathematics, Number 1681. ISBN 3-540-63970-5. 
  19. ^ George Russell (1959). "The Lydian Chromatic Concept of Tonal Organization for Improvisation". Concept Publishing Company, New York. ISBN 0-9703739-0-2. 
  20. ^ S. Cherne (S. P. Fjestad, ed.) (1997). "The Blue Book of Guitars" (4th edition). Blue Book Publications, Inc., Minneapolis, MN. ISBN 1-886768-10-2. 
  21. ^ Perversion of a quote from Oscar Wilde.

[edit] External links

Persondata
NAME Gunther, Neil J.
ALTERNATIVE NAMES
SHORT DESCRIPTION Physicist,Author
DATE OF BIRTH August 15, 1950
PLACE OF BIRTH
DATE OF DEATH
PLACE OF DEATH
Languages