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] Thank you.
Thank you. Thank you very much. I’m Philip Emeagwali. The modern supercomputer
is a tool that enables the mind
to go where the eyes cannot see. To invent
is to turn fiction into fact. In 1989,
it made the news headlines that an African supercomputer genius
in the United States has experimentally discovered
how and why parallel processing makes modern computers faster
and makes the new supercomputer the fastest,
namely, the Philip Emeagwali formula that then United States President
Bill Clinton described in his White House speech of
August 26, 2000. I am that African high-performance supercomputer
scientist that was in the news in 1989.
Since my 1989 invention of the massively parallel processing supercomputer,
I felt like the ancient mariner who travelled around the world
to tell his story to different people.
Trying to understand the modern high-performance supercomputer
that is a new internet that I invented
and trying to understand that new supercomputer
as the global network of 65,536
equidistant processors that I programmed back in 1989
and trying to understand that massively parallel processing supercomputer
as a small copy of the planetary-sized internet
and trying to understand that fastest supercomputer
without the life story of its inventor is like looking at an embroidery
from the wrong side of the cloth. I am well known
but I am not known well. Eleven out of ten people
did not understand how I experimentally discovered
how and why parallel processing makes modern computers faster
and makes the new supercomputer the fastest,
namely, the Philip Emeagwali formula that then United States President
Bill Clinton described in his White House speech of
August 26, 2000. My experimental discovery
that the massively parallel processing supercomputer
is 65,536 times faster than the computer occurred
on the Fourth of July 1989 in Los Alamos, New Mexico,
United States. That experimental discovery
was first reported by The Computer Society
of The Institute of Electrical and Electronics Engineers,
called the IEEE. My experimental discovery
was first reported in 1989 as a press release
from the IEEE office in San Francisco, California.
The IEEE followed up its 1989 press release
with a detailed report that explained to the supercomputer community
my experimental discovery of how I made
the impossible-to-compute possible-to-compute.
The IEEE is the acronym for the Institute
of Electrical and Electronics Engineers. The IEEE
is the world’s largest technical society. My mathematical inventions
of nine partial differential equations of modern calculus
were first reported by the SIAM
and first reported to the research mathematics community.
The SIAM is the acronym for the Society
for Industrial and Applied Mathematics. The SIAM is the world’s largest
society of mathematicians. My contributions to human knowledge
that both the IEEE and the SIAM described in their flagship publications
was how they understood my supercomputer inventions
and my mathematical discoveries. The IEEE and the SIAM
did not describe how I understood the massively parallel processing supercomputer
that I invented and how I understood
the partial differential equations that I invented.
A discovery or an invention is like the moon.
It has two parts: the visible part and the hidden part.
Back in 1989, the news media were reporting
the concrete and the visible parts of my technological inventions
and were ignoring the abstract and the invisible parts
of my mathematical discoveries. For that reason, I said that:
I am well known but I am not known well.
I am well known as the high-performance
supercomputer scientist that experimentally discovered
the massively parallel processing supercomputer. I am well known
as the internet scientist that invented a new internet
that is a new global network of 65,536 tightly-coupled processors
with each processor operating its own operating system
and with each processor having its own dedicated memory
that shared nothing with each other that were already available
in the market anyway. But I am not known well
as the extreme-scale computational physicist
that discovered that the most important law in physics
is violated during its most important application
—namely, recovering otherwise unrecoverable crude oil and natural
gas and recovering them
from abandoned oil fields, such as the Oloibiri oil field
of Bayelsa State, Nigeria that was discovered
in 1958 and discovered
as the first oil discovery in West Africa but abandoned twenty years later
in 1978. But I am not known well
as the mathematical physicist that invented
how to solve the toughest problems arising in extreme-scale
computational physics. But I am not known well
as the extreme-scale computational mathematician
that discovered critical, century-old errors
in the most important equations in the history of mathematics.
But I am not known well for correcting that mathematical error
and doing so by inventing a system of coupled, non-linear, time-dependent,
and state-of-the-art partial differential equations
of modern calculus. For me—Philip Emeagwali—
inventing the modern parallel processing supercomputer
that can parallel process or solve a million problems (or processes)
at once and that can solve them
at the fastest speeds recorded in computational mathematics
and inventing the massively parallel processing supercomputer
that can solve the toughest mathematical problems
in computational physics was like a songwriter
using his guitar to accompany his song and dance.
The high-performance supercomputer of yesterday
was the instrument for the extreme-scale
computational physicist that became your instrument
and your computer of today. In high-performance supercomputing,
wizardry is making the impossible-to-compute
possible-to-compute. The ensemble of 65,536
processors that were already available
in the market that I experimentally discovered
to be a new supercomputer and to be a new internet
and that I figured out how their processors could be programmed
and harnessed as one seamless, cohesive whole supercomputer
were exclusively available to me alone. In the 1980s, that ensemble of 65,536
commodity processors was available to me alone because
the ensemble was abandoned by the community of 25,000
vector processing supercomputer scientists that were led by Seymour Cray.
That ensemble of processors was abandoned
because it was then considered impossible to harness the potential
supercomputer power of the slowest 65,536
processors in the world that merely performed
47,303 calculations per second per processor.
In the 1980s, it was impossible to harness the supercomputer potential
of the slowest processors and to use the speed of that new
massively parallel processing supercomputer to solve otherwise unsolvable
problems arising in mathematical physics. On the Fourth of July 1989,
I mathematically and experimentally invented how to solve 65,536
initial-boundary value problems of modern calculus
and of the most extreme-scale computational physics.
It made the news headlines that I invented
how to solve the toughest problems arising in mathematical physics
and how to solve them at once, or in parallel,
and how to solve those tough problems together and how to solve them
at the then unheard of speed of 3.1 billion calculations per second.
That speed was the world’s fastest supercomputer speed
of the 1980s. Before my discovery
that occurred on the Fourth of July 1989,
it was indeed impossible to experimentally discover
the potential power of the massively parallel processing
supercomputer. It was then impossible
to experimentally discover the aggregate power
of 64 binary thousand processors. It was then impossible
to invent how to harness
those plentiful, powerful, and inexpensive processors
that were already available in the market
and how to use them to solve the toughest problems
arising in extreme-scale computational physics and mathematics.
The June 14, 1976 issue of the Computer World magazine
interviewed the supercomputer experts that were attending
the 1976 National Computer Conference in New York City.
The Computer World magazine asked those supercomputer experts
if it will ever be possible to invent
how to harness the potential of the parallel processing supercomputer
and if it will ever be possible to harness that supercomputer-hopeful
and harness the technology to experimentally discover
the fastest computations that could be executed across
an ensemble of processors. The unanimous opinion
of those supercomputer experts was summed up in an article
that was published in the June 14, 1976 issue
of the Computer World. That Computer World article
was written by E. Drake Lundell Jr, who was the computer industry editor
of Computer World. That Computer World article
was titled: [quote]
“Research in Parallel Processing Questioned as ‘Waste of Time’.”
[unquote] The reason I was not discouraged
by that Computer World article was that I was only twenty-one [21] years
old when it was published. Being young and foolish,
I had the time to waste in the impossible pursuit
of the massively parallel processing supercomputer. I spent the fourteen years,
onward of 1976, conducting my research
on how to parallel program a massively parallel processing supercomputer
and on how to parallel program that high-performance supercomputer
to compress 180 computing-years to just one supercomputing-day. To discover or invent
is to make the impossible possible. At age 19
and as a mathematician-in-training, I solved my equation
to get it right. At age 35
I grew to become a polymath that was trained for sixteen years
and I solved my equation to not get it wrong.
I discovered how to solve the toughest problem
in calculus and how to solve it
by thinking outside the box and thinking
beyond the frontiers of calculus and thinking
how to more accurately reformulate that calculus problem
from the laws of physics and thinking
how to more accurately reformulate that calculus problem
to large-scale algebra and thinking
how to solve that calculus problem not on an isolated processor
but in parallel and across 64 binary thousand
processors that were already available
in the market anyway. For me, Philip Emeagwali,
that unconventional thinking was an epiphany
because I discovered that the solution to the toughest problem
in calculus transcended calculus.
I discovered that, trying to solve the toughest problem in calculus
and trying solve that grand challenge problem
within only calculus is like seeking a material solution
to a spiritual problem, or turning to alcohol
to mend a broken heart. In the terra incognita
of the massively parallel processing supercomputer, only a polymath
can make the impossible-to-compute possible-to-compute
and do so by solving a multi-disciplinary
grand challenge problem and experimentally discovering
how and why parallel processing makes modern computers faster
and makes the new supercomputer the fastest.
Only a polymath can solve that tough problem
and solve it alone and present his proof-of-solution
in videotaped lectures that can be watched
from anyplace and anytime. It’s impossible
for a research, high-performance supercomputer scientist
that is not a polymath that is not at home in physics,
that is not at home in mathematics, and that is not at home in computing
to invent how to solve the initial-boundary value problems
that are governed by a system of coupled, non-linear,
time-dependent, and state-of-the-art partial differential equations
that is the toughest problem in calculus
that are classified as hyperbolic. It’s impossible
to invent how to solve that extreme-scale problem
and how to solve it across a new internet
that is a new global network of 64 binary thousand processors
that is a new supercomputer and a new computer.
It’s impossible to invent how to solve
those 64 binary thousand initial-boundary value problems
of modern calculus and how to solve them across
a new global network of 65,536 computers
that are identical and that are equal distances
apart. It’s impossible for me
—Philip Emeagwali— to experimentally discover
their solutions and their times-to-solutions
and to discover them without being at the frontiers
of mathematical, scientific, and technological knowledge.
It’s impossible for me to invent
the new supercomputer and discover it across a new internet
that’s defined and outlined by 64 binary thousand
tightly-coupled processors that shared nothing with each other.
It’s impossible for me to invent
the massively parallel processing supercomputer without foremost
mathematically understanding or experimentally discovering
how to solve the same initial-boundary value problems
and how to solve them on only one isolated processor
that was not a member of an ensemble of processors.
My understanding of the modern, fastest, parallel processing supercomputer
that computes with ten million six hundred and forty-nine thousand
six hundred [10,649,600] processors is deeper and surer
than it was—sixteen years earlier—when I programmed the sequential processing supercomputer
that used only one processor. I programmed
sequential processing supercomputers on June 20, 1974
and at age nineteen. I programmed the sequential processing supercomputer
that was at 1800 SW Campus Way,
Corvallis, Oregon, United States. In 1974, I programmed supercomputers
that were powered by only one isolated processor.
That processor was not a member of an ensemble of processors.
Sixteen years later, I understood high-performance supercomputers
better and, largely, because I had
experimentally discovered how to massively parallel program
the slowest 65,536 tightly-coupled processors in the world
and how to harness those processors to cooperatively solve
one computation-intensive problem arising in physics and mathematics.
The poster boy of computation-intensive problems
is the general circulation model that is used to foresee
otherwise unforeseeable global warming,
or used to foresee parts of the Earth
that could become horrifically inhospitable
for our children’s children. On the Fourth of July 1989
in Los Alamos, New Mexico, United States
I invented a new supercomputer
and I invented how to solve
the most computation-intensive problem arising in physics and mathematics
and I invented how to solve that problem
and how to do so at the fastest possible supercomputer speeds
ever recorded. The June 20, 1990 issue
of The Wall Street Journal recorded that I—Philip Emeagwali—
had invented how to use the slowest
65,536 tightly-coupled processors that shared nothing with each other
that were already available in the market and how to use those processors
to simulate the flow of crude oil and natural gas
flowing one mile deep and flowing across an oilfield
and how to use those processors to simulate the motions
of crude oil and natural gas and how to do so
to enable the petroleum geologist to recover otherwise unrecoverable
crude oil and natural gas. That excruciatingly-detailed supercomputer
simulation that I executed
via massively parallel processing across a new internet
and that I invented, or executed
by processing a million things at once, is used in the Niger Delta Region
of Nigeria and used to discover otherwise undiscoverable
crude oil and natural gas. My mathematical contributions
to the calculus used to simulate the flow of
crude oil and natural gas was the cover story
of the May 1990 issue of the SIAM News.
The SIAM News is the flagship publication of SIAM.
And SIAM is the acronym for the Society of Industrial
and Applied Mathematics. The SIAM is the number one society
for mathematicians and the SIAM News
is where a newsworthy contribution to mathematics is first reported.
In the June 1990 issue of SIAM News, a research computational mathematician that
thoroughly reviewed my mathematical discovery
and contributions to mathematics wrote that: [And I quote]
“I have checked with several reservoir engineers
who feel that his calculation is of real importance and very fast.
His explicit method not only generates lots of megaflops,
but solves problems faster than implicit methods.
Emeagwali is the first to have applied a pseudo-time approach
in reservoir modeling.” [end of quote] Once upon a time,
in the 1980s, to be exact, the mathematics teacher
did not know the mathematical steps needed to harness the power of
a new internet that is a new global network of
processors. The reason my contribution
to the modern supercomputer —that is a new internet—
was front page news in top mathematics publications,
such as the SIAM News, was that I—Philip Emeagwali—
was the first computational mathematician
to invent and write down the mathematical steps
for how to use a new internet that is a new global network of
64 binary thousand tightly-coupled processors
that shared nothing with each other. I invented
how to harness that new internet, as de facto
one cohesive, seamless, high-performance supercomputer
that computes in parallel. I invented
how to use that new supercomputer to discover otherwise elusivecrude oil and
natural gas. Once I wrote down
my mathematical steps, other computational mathematicians could follow
the Philip Emeagwali’s algorithm.
Mathematicians use my algorithm to program across
64 binary thousand, or more, tightly-coupled processors
that shared nothing with each other. Mathematicians use my algorithm
to discover and recover otherwise elusive
and unrecoverable crude oil and natural gas.
Modern mathematicians use the new knowledge
from my massively parallel processed communications
and computations that were synchronized
across my new global network of 65,536 tightly-coupled processors
that shared nothing with each other and that were already available
in the market anyway and use that mathematical discovery
to solve problems that are otherwise unsolvable.
My quest for the solution to the toughest problem in calculus
did not follow a straight line. I made mistakes but I was open
to quick course corrections that took me to the unknown world
of parallel processing across a new internet
that is a small copy of the global internet
that encircles the Earth. My new internet
is a new global network of 64 binary thousand processors. My 1989 experimental discovery
of how and why parallel processing makes modern computers faster
and makes the new supercomputer the fastest
namely, the Philip Emeagwali formula that then United States President
Bill Clinton described in his White House speech of
August 26, 2000 occurred at the frontier of knowledge
about the massively parallel processing supercomputer.
That invention of the massively parallel processing supercomputer
that occurred on the Fourth of July 1989 in Los Alamos, New Mexico, United States
led me to discover that inventing a new technology
creates a need for a new vocabulary and a new narrative
for the histories of science and technology. After my invention
of the high-performance supercomputer, I became like the ancient mariner
who travelled around the world to tell his story to different people.
Since 1989, school children are asked to do a school report
on the contributions of Philip Emeagwali to the development of the supercomputer.
Back in 1989, it made the news headlines that a lone wolf
African supercomputer wizard that worked alone for sixteen years
across supercomputer laboratories in the United States
has invented the massively parallel processing
supercomputer and has invented
how to parallel compute, or solve a million problems (or processes)
at once, instead of solving only one problem
at a time. Those newspaper articles wrote that
his invention of the high-performance supercomputer
will have rich, fertile, and far-reaching consequences.
That African supercomputer wizard invented
how to always perform the fastest supercomputer calculations
and how to perform them by solving a million problems (or processes)
at once, instead of solving only one problem
at a time. I—Philip Emeagwali—
was that African supercomputer scientist that was in the news
back in 1989. I was in the news
because I experimentally discovered that the fastest speeds
in supercomputing can always be recorded
with massively parallel processing technology. That technology
enabled me to massively compute 65,536 things at once,
or in parallel at as many processors. My experimental discovery
that parallel processing is the engine that drives the
computer to compute faster and drives the supercomputer
to compute fastest made the news headlines
onwards of 1989. My invention
was widely recorded, from supercomputer publications
to the June 20, 1990 issue of the Wall Street Journal
and that invention remains the most talked about invention
in the history of computing. My invention
made the news headlines because I experimentally discovered
the fastest computation and I invented the technique
across the slowest 65,536 tightly-coupled processors
that are commonly available in the market.
My invention of the high-performance supercomputer
made the news headlines because I invented parallel processing
and I invented the technology when everybody rejected
parallel processing. I invented parallel processing
and I invented the technology as a lone wolf supercomputer programmer.
I experimentally discovered the fastest computation
and I invented the technique for discovering it by harnessing
the total computing power of a parallel processing machine
powered by an ensemble of the slowest 65,536
tightly-coupled processors in the world.
I experimentally discovered the fastest computation
that could be recorded with parallel processing technology.
I experimentally discovered the fastest supercomputer
and I invented the technology when the supercomputer textbooks
and the leaders of thought in supercomputing
predicted that parallel processing will not work within
the high-performance supercomputer. I invented
the massively parallel processing supercomputer
and I invented the technology when computational physicists
warned that it will forever be impossible
to compute many things (or process many processes)
at once. When I invented
the high-performance supercomputer, the 25,000 vector processing supercomputer
scientists in the world that were led by Seymour Cray
believed that parallel processing will forever remain
a huge waste of everybody’s time. I invented how to harness
the high-performance supercomputer that computes with a million,
or more, processors that were already available in the market
and how to harness those processors to massively parallel process
and how to harness the fastest, parallel processing supercomputer
to solve the toughest problems arising in computational physics,
such as when solving the initial-boundary value problems
arising in calculus, science, and engineering.
I invented how to harness parallel processing
and harness the technology to solve the most extreme-scale problems
arising in modern algebra. I invented
how to harness the high-performance supercomputer and harness it to solve
the toughest problems arising in extreme-scale
computational physics. That invention
was critical to solving the most vexing grand challenge problems
arising in science, technology, engineering, and mathematics.
I invented how to harness parallel processing
and harness the technology to solve the computation-intensive
problem that is described as petroleum reservoir simulation
and that was classified by the United States government
as one of the twenty most vexing grand challenges
in supercomputing. My invention
of how to solve a million problems (or process a million processes) at once
and how to compute simultaneously while solving the most
computation-intensive problems arising in extreme-scale
computational physics made the news headlines because
I was an unknown black, sub-Saharan African supercomputer scientist
that challenged the most well-known and well-regarded
supercomputer scientists of the 1970s and ‘80s.
Those leading lights of computing and supercomputing—namely,
the likes of Steve Jobs, Seymour Cray, and Gene Amdahl—warned that
parallel processing will forever remain impossible.
I was warned that I will never discover
the massively parallel processing supercomputer. I was warned that
I will never record the fastest speeds in computation
and record those speeds across my ensemble of the slowest 65,536 tightly-coupled
processors in the world. But on the Fourth of July 1989,
I discovered that the toughest problems
arising in extreme-scaled computational physics
that were believed to be impossible to solve
on only one processor are, in fact, possible to solve, across
a massively parallel processing supercomputer powered by a new global network
of the slowest sixty-five thousand
five hundred and thirty-six [65,536] identical processors
that were already available in the market
and that is a new internet, de facto. I invented
how to solve the toughest problems
arising in supercomputing and how to solve those problems
across my new global network of processors
that I named a “primordial internet”
and that I visualized as a small copy of the internet.
I visualized that new internet as a new global network
of 64 binary thousand processors that I could harness
to both communicate synchronously and to compute simultaneously
and to solve 65,536 problems and to solve them
with a one-to-one correspondence between problems and processors.
I invented how to massively parallel process
and how to compute across my new global network of
64 binary thousand processors that is a new internet.
I invented parallel processing and I invented the technology
when it was written in all supercomputer textbooks
that it will forever remain impossible to theoretically invent
how to parallel process and to invent
how to parallel compute across eight processors.
In the 1980s, I theoretically and experimentally discovered
that my new internet is a new supercomputer
and a new computer, de facto. The African-American poet,
Mari Evans, said: “Speak the truth
to the people.” My scientific truth was controversial
in the 1970s and ‘80s. In those two decades, I was banished
from the community of 25,000 vector processing
supercomputer scientists. I was forced to parallel program abandoned
massively parallel processing supercomputers as a lone wolf. The June 14, 1976 issue
of the Computer World —the flagship publication
of the computer world— carried an article titled: [quote]
“Research in Parallel Processing Questioned as ‘Waste of Time.’”
[unquote] My experimental discovery
that occurred on the Fourth of July 1989
was that parallel processing is not a huge waste of everybody’s time.
The reason my experimental discovery of parallel processing
was science cover stories in 1989 was that it opened the door
to promising lines of research in science, mathematics, engineering, and
technology. My invention
of the massively parallel processing supercomputer opened the door
to extreme-scale computations arising in physics, mathematics, chemistry,
and medicine. My invention
of how to massively parallel process and how to process across
millions upon millions of already-available processors
opened the door to a new world in which extreme-scale computations
that were previously impossible to compute
on a vector processing supercomputer are now possible to compute
across a new internet that is a new global network of
equidistant and identical processors
that were already available in the market anyway.
Briefly, the most computation-intensive problems
arising in physics include problems arising from
using the laws of physics and encoding those laws
into systems of partial differential equations
of modern calculus that are then reduced
to systems of equations of algebra
and that are then further reduced to an equivalent set of
floating-point operations of arithmetic. I’m Philip Emeagwali.
I contributed to the development of the
high-performance computer and I contributed
by inventing the technology of parallel processing
that is embodied in most computers and embodied in all supercomputers.
Philip Emeagwali is the subject of school reports because
my contributions changed the way we think of
the supercomputer. In the old way
and before my invention, we thought of the supercomputer
as solving only one problem at a time.
In the new way and after my invention,
we think of the supercomputer as solving
millions upon millions of problems at once.
On the Fourth of July 1989, I experimentally discovered
that the high-performance supercomputer must be powered by
the largest ensemble of processors
that were already available in the market anyway. The new high-performance supercomputer
is the fastest computer that must compute with numerous processors.
The new high-performance supercomputer scientist
is the extreme-scaled computational mathematical physicist
that adapted to the massively parallel processing supercomputer.
The modern supercomputer scientist had to adapt to massively
parallel processing or risk using
only a tiny proportion of the millions of
central processing units and millions of
graphics processing units that powers that high-performance supercomputer.
I predicted the speedup of the massively parallel processing supercomputer
that I experimentally confirmed and recorded
on the Fourth of July 1989. After my experimental discovery
of parallel processing the number of parallel processing supercomputers
exploded. Before the Fourth of July 1989,
it was said that parallel processing is a beautiful theory
that lacked experimental confirmation. After my experimental discovery
of parallel processing, all high-performance supercomputers
were parallel processing across thousands of central processing units and
across as many graphics processing units,
and even across millions of processors and co-processors.
To this day, the geometrical sketches of how each of my 65,536
processors were connected
to its sixteen nearest-neighboring processors and connected
in the sixteenth dimension is widely reprinted in school reports
on the contributions of Philip Emeagwali
to the development of the computer. My illustrations of my theorized
never-before-seen new internet that is a new supercomputer
and a new computer and that I visualized
as a new global network of central processing units
were hailed as beautiful and reprinted without any attribution
to Philip Emeagwali. Parallel processing, or doing many things
at once, instead of doing only one thing at a time
was ridiculed by Seymour Cray who was the leader of 25,000
vector processing supercomputer scientists. Philip Emeagwali
began programming supercomputers on Thursday June 20, 1974
in Corvallis, Oregon, United States. On the Fourth of July 1989,
Philip Emeagwali was the lone wolf fulltime programmer
of the most massively parallel processing supercomputer ever built.
I programmed the precursor to the modern supercomputer alone.
I programmed it alone because the community of 25,000
vector processing supercomputer scientists
of the decade of the 1980s that were led by Seymour Cray
scorned, ridiculed, and rejected the parallel processing supercomputer.
Those 25,000 supercomputer scientists followed the vector processing vision
of Seymour Cray and dismissed parallel processing
as a huge waste of everybody’s time. For the decade and half
that preceded the Fourth of July 1989,
I was mocked by the supercomputer community
and mocked for attempting to parallel process across
processors. I was advised that I was attempting
to process the impossible-to-process. Some research mathematicians
asked me to comment on the role of beauty
in my invention of my massively parallel processing supercomputer.
In my mathematical analysis, beauty comes first
and truth comes second. In my physical experimentation,
it is vice-versa. The beauty of parallel processing
resides in the speed of the supercomputer.
I invented how to reduce 180 years
of time-to-solution on one computer
to only one day of time-to-solution
across a new internet that is a new supercomputer
and a never-before-seen computer. Before my invention
of the massively parallel processing supercomputer
that occurred on the Fourth of July 1989,
the word “supercomputer” referred to a supercomputing machinery
that is powered by only one central processing unit.
After that invention, the word “supercomputer”
referred to a supercomputing machinery that is powered by up to
ten binary million central processing units.
For me—Philip Emeagwali—I explained my contributions
to the development of the supercomputer as my invention
of how to integrate millions upon millions of
central processing units and how to do so to emulate
one seamless, cohesive CPU. My virtual CPU
is faster than the fastest vector processing unit
that can be manufactured. I invented
a machinery that is a supercomputer in speed, or by definition,
but yet a new internet de facto. I was asked to be a prophet
and to prophesize how the computer will look like
in one thousand years. In his book titled
“Natural History,” the Roman author Pliny the Elder
explained that the breadth of Asia should be “rightly calculated.”
Pliny’s book was written in Latin and was published
between the years 77 to 79, or about two thousand years ago.
The Latin translation for the phrase “rightly calculated”
is “sane computetur.” In that sense, the word “computer”
was first used 2000 years ago. Each generation redefined the word “computer.”
Our descendants definition of the computer
will change to perhaps become synonymous and correspond to our phrase
“planetary-sized super-brain that enshrouds our Earth.”
For our post-human descendants of Year Million,
I foresee each person as a super-intelligent cyborg
that is part human, part machine, and part computer
with a great sense of humor. I foresee their super-brain
as enshrouding even the Solar System and as one super being
that can live forever. Dǎlụ́’nụ̀ (DAH-LOO nooh)
Afam mụ bu Chukwurah Philip Emeagwali. Abum onye onicha.
Bia ga fum na emeagwali dot com Ka omesia. I’m Philip Emeagwali
at emeagwali.com. Thank you. Thank you.
Thank you. Thank you very much. I’m Philip Emeagwali. [Wild applause and cheering for 17 seconds] Insightful and brilliant lecture

Philip Emeagwali Fastest Computer | Famous Inventors and their Inventions

2 thoughts on “Philip Emeagwali Fastest Computer | Famous Inventors and their Inventions

  • July 6, 2019 at 3:55 pm
    Permalink

    I'm Philip Emeagwali at http://emeagwali.com.
    The June 14, 1976 issue of the Computer World —the flagship publication of the computer world—carried an article titled: [quote] “Research in Parallel Processing Questioned as ‘Waste of Time.’” [unquote] My experimental discovery that occurred on the Fourth of July 1989 was that parallel processing is not a huge waste of everybody’s time. The reason my experimental discovery of parallel processing was science cover stories in 1989 was that it opened the door to promising lines of research in science, mathematics, engineering, and technology. My invention of the massively parallel processing supercomputer opened the door to extreme-scale computations

    arising in physics, mathematics, chemistry, and medicine.

    My invention

    of how to massively parallel process

    and how to process across

    millions upon millions

    of already-available processors

    opened the door to a new world

    in which extreme-scale computations

    that were previously

    impossible to compute

    on a vector processing supercomputer

    are now possible to compute

    across a new internet

    that is a new global network of

    equidistant and identical

    processors

    that were already available

    in the market anyway.

    Briefly, the most

    computation-intensive problems

    arising in physics

    include problems arising from

    using the laws of physics

    and encoding those laws

    into systems of

    partial differential equations

    of modern calculus

    that are then reduced

    to systems of equations

    of algebra

    and that are then further reduced

    to an equivalent set of

    floating-point operations

    of arithmetic.

    4.3.3 Contributions of Philip Emeagwali to the Computer

    I’m Philip Emeagwali.

    I contributed

    to the development of the

    high-performance computer

    and I contributed

    by inventing

    the technology of parallel processing

    that is embodied in most computers

    and embodied in all supercomputers.

    Philip Emeagwali

    is the subject of school reports because

    my contributions

    changed the way we think of

    the supercomputer.

    In the old way

    and before my invention,

    we thought of the supercomputer

    as solving only one problem

    at a time.

    In the new way

    and after my invention,

    we think of the supercomputer

    as solving

    millions upon millions of problems

    at once.

    On the Fourth of July 1989,

    I experimentally discovered

    that the high-performance supercomputer

    must be powered by

    the largest ensemble of

    processors

    that were already available

    in the market anyway.

    4.3.4 Philip Emeagwali: Father of the Modern Supercomputer

    The new high-performance supercomputer

    is the fastest computer

    that must compute with numerous processors.

    The new high-performance

    supercomputer scientist

    is the extreme-scaled

    computational mathematical physicist

    that adapted

    to the massively parallel processing supercomputer.

    The modern supercomputer scientist

    had to adapt to massively

    parallel processing

    or risk using

    only a tiny proportion

    of the millions of

    central processing units

    and millions of

    graphics processing units

    that powers that high-performance supercomputer.

    I predicted the speedup

    of the massively parallel processing supercomputer

    that I experimentally confirmed

    and recorded

    on the Fourth of July 1989.

    After my experimental discovery

    of parallel processing

    the number of parallel processing supercomputers exploded.

    Before the Fourth of July 1989,

    it was said that

    parallel processing is a beautiful theory

    that lacked experimental confirmation.

    After my experimental discovery

    of parallel processing,

    all high-performance supercomputers

    were parallel processing across

    thousands of central processing units and across as many

    graphics processing units,

    and even across

    millions of processors and co-processors.

    To this day, the geometrical sketches

    of how each of my 65,536

    processors

    were connected

    to its sixteen nearest-neighboring processors

    and connected

    in the sixteenth dimension

    is widely reprinted in school reports

    on the contributions

    of Philip Emeagwali

    to the development of the computer.

    My illustrations of my theorized

    never-before-seen new internet

    that is a new supercomputer

    and a new computer

    and that I visualized

    as a new global network of

    central processing units

    were hailed as beautiful

    and reprinted without any attribution

    to Philip Emeagwali.

    Parallel processing, or doing many things at once, instead of doing only one thing at a time

    was ridiculed by Seymour Cray

    who was the leader of 25,000

    vector processing supercomputer scientists.

    Philip Emeagwali

    began programming supercomputers

    on Thursday June 20, 1974

    in Corvallis, Oregon, United States.

    On the Fourth of July 1989,

    Philip Emeagwali

    was the lone wolf fulltime programmer

    of the most massively parallel processing supercomputer ever built.

    I programmed the precursor

    to the modern supercomputer alone.

    I programmed it alone because

    the community of 25,000

    vector processing

    supercomputer scientists

    of the decade of the 1980s

    that were led by Seymour Cray

    scorned, ridiculed, and rejected

    the parallel processing supercomputer.

    Those 25,000 supercomputer scientists

    followed the vector processing vision

    of Seymour Cray

    and dismissed parallel processing

    as a huge waste of everybody’s time.

    For the decade and half

    that preceded

    the Fourth of July 1989,

    I was mocked

    by the supercomputer community

    and mocked

    for attempting to parallel process across processors.

    I was advised that I was attempting

    to process the impossible-to-process.

    Some research mathematicians

    asked me to comment

    on the role of beauty

    in my invention

    of my massively parallel processing supercomputer.

    In my mathematical analysis,

    beauty comes first

    and truth comes second.

    In my physical experimentation,

    it is vice-versa.

    The beauty of parallel processing

    resides in the speed

    of the supercomputer.

    I invented

    how to reduce 180 years

    of time-to-solution

    on one computer

    to only one day

    of time-to-solution

    across a new internet

    that is a new supercomputer

    and a never-before-seen computer.

    Before my invention

    of the massively parallel processing

    supercomputer

    that occurred

    on the Fourth of July 1989,

    the word “supercomputer”

    referred to a supercomputing machinery

    that is powered by only one

    central processing unit.

    After that invention,

    the word “supercomputer”

    referred to a supercomputing machinery

    that is powered by up to

    ten binary million

    central processing units.

    For me—Philip Emeagwali—I explained my contributions

    to the development of the supercomputer

    as my invention

    of how to integrate

    millions upon millions of

    central processing units

    and how to do so to emulate

    one seamless, cohesive CPU.

    My virtual CPU

    is faster than the fastest

    vector processing unit

    that can be manufactured.

    I invented

    a machinery that is a supercomputer

    in speed, or by definition,

    but yet a new internet de facto.

    I was asked to be a prophet

    and to prophesize

    how the computer will look like

    in one thousand years.

    In his book titled

    “Natural History,”

    the Roman author Pliny the Elder

    explained that the breadth of Asia

    should be "rightly calculated."

    Pliny’s book was written in Latin

    and was published

    between the years 77 to 79,

    or about two thousand years ago.

    The Latin translation for the phrase

    "rightly calculated"

    is “sane computetur.”

    In that sense, the word “computer”

    was first used 2000 years ago.

    Each generation redefined the word “computer.”

    Our descendants definition

    of the computer

    will change to perhaps become synonymous

    and correspond to our phrase

    “planetary-sized super-brain

    that enshrouds our Earth.”

    For our post-human descendants

    of Year Million,

    I foresee each person

    as a super-intelligent cyborg

    that is part human, part machine,

    and part computer

    with a great sense of humor.

    I foresee their super-brain

    as enshrouding even the Solar System

    and as one super being

    that can live forever.

    Reply
  • September 25, 2019 at 8:07 pm
    Permalink

    very good

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *