Philip Emeagwali

From Wikipedia, the free encyclopedia

Philip Emeagwali (born in 1954) is a Nigerian-born computer scientist/geologist who was one of two winners of the 1989 Gordon Bell Prize, a prize from the IEEE, for his use of the Connection Machine supercomputer – a machine featuring over 65,000 parallel processors – to help analyze petroleum fields.

1 Biography
2 Awards
3 Court case
4 Claims
5 References
6 External links

[edit] Biography

Emeagwali was born in Akure, Nigeria in 1954.^[1] He dropped out of school in 1967 because of the Nigerian civil war. When he turned fourteen, he was conscripted into the Biafran army. After the war he completed a high-school equivalency through self-study and came to the United States to study at university under a scholarship. He received a bachelor's degree in mathematics from Oregon State University in 1977. He received a master's degree in environmental engineering from George Washington University in 1981, and another master's degree in Mathematics from the University of Maryland, College Park in 1986. He also received a post-master's degree in ocean, coastal and marine engineering from George Washington University in that year. He was also working as a civil engineer at the Bureau of Land Reclamation in Wyoming during this period.

[edit] Awards

Emeagwali received the $1,000^[2] 1989 Gordon Bell Prize, based on an application of the CM-2 massively-parallel computer for oil-reservoir modeling. He won in the "price/performance" category, with a performance figure of 400 Mflops/$1M, corresponding to an absolute performance of 3.1 Gflops. (The winning entry in the "peak performance" category that year – coincidentally also for oil-related seismic data processing on a CM-2 – actually achieved 6 Gflops, or 500 Mflops/$1M, but the judges decided not to award both prizes to the same team.)^[3] This simulation was the first program to apply a pseudo-time approach to reservoir modeling.^[4]

Apart from the prize itself, there is no evidence that Emeagwali's work was ever accepted for publication in the peer-reviewed scientific literature, nor that it had any other lasting impact on the field of high-performance computing.^[5] Nevertheless, based on the Bell Prize, he has since received numerous further awards^[6], ranging from one from the World Bank-IMF Africa Club to being voted the "35th-greatest African of all time" in a survey by New African magazine.^[7] His achievements were quoted in a speech by Bill Clinton as an example of what Nigerians could achieve when given the opportunity.^[8]

[edit] Court case

Emeagwali studied for a Ph.D. degree from the University of Michigan from 1987 through 1991. His thesis was not accepted by a committee of internal and external examiners and thus he was not awarded the degree. Emeagwali filed a court challenge, claiming that the decision was a violation of his civil rights and that the university had discriminated against him in several ways because of his race. The court challenge was dismissed, as was an appeal to the Michigan state Court of Appeals.^[9]

[edit] Claims

This article or section needs sources or references that appear in reliable, third-party publications.
Primary sources and sources affiliated with the subject of the article are generally not sufficient for a Wikipedia article. Please include more appropriate citations from reliable sources, or discuss the issue on the talk page.
This article has been tagged since December 2007.

In interviews^[10]:

Emeagwali says he formulated new^{[citation needed]}^[10] partial differential equations for moving liquids and gases such as oil and natural gas flow within the Earth's interior.
Emeagwali says he proved^{[citation needed]}^[10] that the use of only Dirichlet type boundary conditions yields more accurate numerical solutions near petroleum production wells.
Emeagwali says he derived error stopping criterion for computational fluid dynamics software.^{[citation needed]}^[10]
Emeagwali says he, in 1988, programmed a Connection Machine with 2,048 processors, using those equations and criterion, to simulate an oil reservoir.^[10]
- Emeagwali says this simulation implemented the first^{[citation needed]}^[10] petroleum reservoir model on a massively parallel computer.
- Emeagwali says this simulation performed the world's fastest^{[citation needed]}^[10] computation of 3.1 billion calculations per second in 1988.
- Emeagwali says this simulation was the first program for a massively parallel computer that outperformed equivalent programs for conventional (vector) supercomputers,^{[citation needed]}^[10] proving that it is possible to efficiently harness thousands of parallel processors. (In the 1980's, it was generally believed^[who?] that Amdahl's law would give such diminishing returns that it would be pointless to use more than a handful of parallel processors -- it was generally believed^[who?] that a few large, expensive, fast, serial processors would cost less and have higher performance than lots of low-cost parallel processors. In particular, Seymour Cray famously quipped, "If you were plowing a field, which would you rather use? Two strong oxen or 1024 chickens?"[1])
- Emeagwali says this simulation solved^{[citation needed]}^[10] one of America's 20 Grand Challenges: accurately computing how oil flows underground
Emeagwali says he wrote a 1000-page report describing the new equations and how Emeagwali implemented them on the Connection Machine.^{[citation needed]}^[10]
Emeagwali says he told the petroleum industry that they could recover more oil by using massively parallel computer simulations^[10]
Emeagwali says he discovered the analogy between Darcy's equations used in petroleum reservoir simulations and geostrophic equations used in weather forecasting.^{[citation needed]}^[10]
Emeagwali says he programmed the largest weather forecasting simulation (128 million points) in 1990 on a Connection Machine with 65,536 processors^{[citation needed]}^[10]
- Emeagwali says this simulation set world record for parallel computer speedup of 65536 in 1990,^{[citation needed]}^[10] experimentally demonstrating that thousands of parallel processors can be efficiently harnessed. Today this fact is common knowledge, known as scalability.

Emeagwali says he formulated the counter-intuitive speedup paradox which states that there are two different theoretical speedup of parallel computers, and proved that both answers are right.^{[citation needed]}^[10]
Emeagwali says he discovered the counter-intuitive hypercube paradox.^{[citation needed]}^[10]
Emeagwali says he formulated the theory of weak nearest-neighbors in parallel computing.^{[citation needed]}^[10]
Emeagwali says he demonstrated that the most communication-efficient parallel programs must be computation-inefficient.^{[citation needed]}^[10]
Emeagwali says he introduced the concept of network frequency for parallel computers.^{[citation needed]}^[10]
Emeagwali says he discovered chirality in parallel programming.^{[citation needed]}^[10]
Emeagwali says he discovered enantiomeric networks.^{[citation needed]}^[10]
Emeagwali says he discovered the relationship between sphere packing and fast computing.^{[citation needed]}^[10]