Greatest Computational Mathematicians in History | Philip Emeagwali
TIME magazine called him
“the unsung hero behind the Internet.” CNN called him “A Father of the Internet.”
President Bill Clinton called him “one of the great minds of the Information
Age.” He has been voted history’s greatest scientist
of African descent. He is Philip Emeagwali.
He is coming to Trinidad and Tobago to launch the 2008 Kwame Ture lecture series
on Sunday June 8 at the JFK [John F. Kennedy] auditorium
UWI [The University of the West Indies] Saint Augustine 5 p.m.
The Emancipation Support Committee invites you to come and hear this inspirational
mind address the theme:
“Crossing New Frontiers to Conquer Today’s Challenges.”
This lecture is one you cannot afford to miss. Admission is free.
So be there on Sunday June 8 5 p.m.
at the JFK auditorium UWI St. Augustine. [Wild applause and cheering for 22 seconds] [My Struggles to Invent New Mathematics] My laborious Taylor’s expansions
of 1981 were how I approximated
the value of each of my solution by taking the sum of its derivatives
at a given point. Taylor’s expansions yielded
my a priori error estimates that I used to pre-select
the most, hopefully, accurate finite difference algebraic approximations
of the nine partial differential equations, called Philip Emeagwali’s Equations,
that I invented. I contributed to mathematical knowledge
and my contribution to algebra and calculus was the cover story
of top mathematics publications, such as the June 1990 issue
of the SIAM News. One fact that I never mentioned before
was that I often pursued false mathematical trails.
Back in 1981, I was unreasonably obsessed
with the Hopscotch algorithms as a numerical solution of
partial differential equations. I was obsessed with Hopscotch methods
because I was unreasonably optimistic and believed that
Hopscotch methods are hybrid explicit-implicit methods
that could be very accurate and that Hopscotch methods
could enable me to email my initial-boundary value problems
and email them across a new internet
that I visualized as a new global network of
65,536 commodity-off-the-shelf processors. After a year of seemingly endless mathematical
analyses of Hopscotch algorithms
and computational experiments of Hopscotch
computational fluid dynamics codes I discovered that
I was following a false trail and that hopscotch algorithms
were over hyped. After wasting extraordinary amount of time,
I resettled on explicit finite difference approximations.
In the end, I invented explicit finite difference
algebraic approximations of the nine partial differential equations,
called Philip Emeagwali’s Equations, that I contributed to modern calculus.
That was how I scribbled new calculus that had never been scribbled
on any blackboard before. That was how I coded new algebra
that had never been coded by any computational algebraist before.
That was how I saw a new supercomputer that had never been seen
by any supercomputer scientist before. [Father of Large-Scale Algebra] It’s often said that
parallel processing across millions upon millions
of tightly-coupled commodity-off-the-shelf processors
that shared nothing with each other is the biggest advance in computing
since the programmable computer was invented
back in 1946. In my country of birth, Nigeria,
a million billion trillion floating-point arithmetical computations
are massively parallel processed each day
and massively parallel processed to discover and recover
the otherwise elusive crude oil and natural gas
that are buried a mile deep in the Niger-Delta oilfields of Nigeria.
As a discoverer-hopeful, back in 1974, in Corvallis, Oregon, United States,
I asked a big question that had never been answered before.
That overarching question was: “How do we parallel process
across a new internet that is a new global network of
64 binary thousand computers?” If that big question
that I asked in 1974 was already answered,
or if parallel processing was already discovered,
my invention of the massively parallel processing supercomputer
will not have been cover stories and would not have been recorded
in the June 20, 1990 issue of The Wall Street Journal.
If the answer to that big overarching question
was known, I would not have gotten telephone calls
from the likes of Steve Jobs who wanted to know
how I invented the massively parallel processing supercomputer
that is faster than the vector processing supercomputer.
Steve Jobs wanted to know how I recorded 3.1 billion calculations per second.
As an aside, my invention of parallel processing
that occurred on the Fourth of July 1989 inspired Steve Jobs
to use four processors that processed in parallel
to also attain a speed of 3.1 billion calculations per second
and record that speed in his first Apple personal supercomputers,
called the Power Mac G4. Steve Jobs introduced
his personal supercomputer at the Seybold conference
that took place in San Francisco on August 31, 1999.
Like the modern supercomputer, the fastest speed in your computer
are coming from parallel processing. The new supercomputer knowledge
that made the news headlines was that I—Philip Emeagwali—had invented
how to massively parallel process and that I invented the technology
that drives the modern supercomputer and invented the technology
on the Fourth of July 1989 and invented the technology
in Los Alamos, New Mexico, United States. I invented
the parallel processing supercomputer technology
to enable me to solve the toughest problems
arising in extreme-scale algebra. Such mathematical physics problems arise
when trying to discover and recover crude oil and natural gas
and do so from the Niger-Delta oilfields of my country of birth, Nigeria.
My quest for the new algebra that is my contribution to algebra
began with the arithmetic times table that I memorized in 1960
in first grade at Saint Patrick’s Primary School, Sapele,
in the then Western Region of the then British West African colony
of Nigeria. That times table
went to only twelve times twelve. That times table
was near the beginning of knowledge of arithmetic.
On the Fourth of July 1989, in Los Alamos, New Mexico, United States,
I—Philip Emeagwali—mathematically invented how to massively parallel process
arithmetic times tables and parallel process them across
a new internet that is a new global network of processors.
I invented new algorithms, or new instructions,
that told each processor what to compute within itself
and what to communicate to its up to sixteen nearest-neighboring processors.
Since the first programmable supercomputer was invented in 1946,
each supercomputer manufactured was faithful to its primary mission, namely,
to solve the most extreme-scale problems arising in computational physics
and to increase productivity, reduce time-to-solution,
and reduce time-to-market. Supercomputing is mathematics-intensive.
For that reason, most supercomputer scientists are, in part,
research computational mathematicians. In supercomputing
and in computational physics, to discover is to make the impossible-to-solve
possible-to-solve. The first person, or the discoverer,
makes the impossible possible, and thereafter, everybody knows that
parallel processing is no longer a waste of everybody’s time.
I—Philip Emeagwali—was credited for making the invention
of massively parallel processing, the technology that makes supercomputers
fastest. I invented
parallel processing when the supercomputer technology
was scorned, ridiculed, and rejected by the likes of Steve Jobs.
I invented parallel processing
when the supercomputer technology was presumed
to be untestable and even wrong. My discovery
that the impossible-to-solve arising in extreme-scale
algebraic computations is possible-to-solve
across a new internet that is a new supercomputer
and a new computer was recorded in the June 20, 1990 issue
of the Wall Street Journal. I’m Philip Emeagwali.
at emeagwali.com. Thank you. [Wild applause and cheering for 17 seconds] Insightful and brilliant lecture