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] [A Father of the Internet] Thank you. Thank you. Thank you very much. I’m Philip Emeagwali. The Internet
has many fathers and mothers as well as aunts and uncles
that did not invent a new internet. The father of the Internet
should at least invent a new internet. I am called a father of the Internet because
I am the only father of the Internet that invented a new internet. [Philip Emeagwali Internet] [Inventing a New Internet From Science Fiction] Back in 1989,
I was in the news headlines because I was the first person
to discover how to parallel process
real-world problems and how to do so across
millions upon millions of processors that were tightly-coupled to each other.
In parallel supercomputing across my new internet
that was a new global network of 65,536 processors
that were identical to each other the most important knowledge
is to fully understand how to control and harness
every processor that was within my global network of processors.
Each processor operated its own operating system.
In 1989, I made the news headlines when I recorded
the maximum possible speed increase across my ensemble of
65,536 commodity-off-the-shelf processors. That parallel processed speed increase
of a factor of 65,536 that was then considered impossible
led to my discovery that parallel supercomputing
will be the vital technology that will make computers faster
and make supercomputers fastest. I discovered
practical parallel supercomputing and did so
by making a one-to-one corresponded and metaphorical mapping
and doing so from the vertices of the hypercube
to each processor and by making another
one-to-one corresponded mapping from the bi-directional edges
of the hypercube to my email wires. Unlike your cube,
my hypercube was defined in sixteen-dimensional hyperspace
and therefore has two-raised-to-power sixteen,
or 64 binary thousand, vertices and sixteen times as many,
or one binary million, bi-directional edges.
Programming 64 binary thousand tightly-coupled processors
to work together to forecast the weather was in the realm of science-fiction
and would have been dismissed as an act of insanity
and dismissed when I began programming supercomputers
back on June 20, 1974 at 1800 SW Campus Way,
Corvallis, Oregon, United States. But on the Fourth of July 1989
in Los Alamos, New Mexico, United States, I discovered
how to turn that science fiction into non-fiction
that is used to forecast the weather for your evening news.
As the first person to make the news headlines
for discovering practical parallel supercomputing,
I visualized the vertices and the edges of a hypercube
that were etched onto the surface of a hypersphere.
I invented the Philip Emeagwali Internet
as corresponding to the cut-out silhouette
that was my topological metaphor for the appearance of my new internet
that was a new global network of 65,536 processors
that were identical to each other and that were tightly-coupled
to each other. That new supercomputer
and that new internet were like tightly-conjoined twins
with only one superbrain. In 1989, I was in the news headlines because
I discovered practical parallel supercomputing,
and invented it as the vital technology that now underpins every supercomputer
and that enabled me to compute faster and across my new internet
and to compute faster than any supercomputer that ever existed.
I envisioned my new virtual supercomputer
as my new internet. My first visions of my new internet
began as a dark shape and as an outline
of a blank global network of processors that were connected with email wires
that were empty of messages. That dark shape of my new internet remained
visible inside my mind. It’s impossible for me to work alone
and program that new internet and do so without intellectually seeing the
exact positions in sixteen dimensional space
of each of my 65,536 processors. It’s also impossible
for me to send and receive email messages across
my 1,048,576 bi-directional email wires and do so correctly
without knowing, in advance, the exact positions in hyperspace
of my 65,536 processors. As the first
parallel supercomputer scientist, I was not trying to see
with my naked eyes, any of my 65,536 processors
or any of my 1,048,576 bi-directional email wires
that married my commodity processors together and did so to form a new internet
that tightly-circumscribed a globe in sixteen-dimensional hyperspace.
In contrast, I only saw my new supercomputer
and my new internet inside my mind, not with my naked eyes
as was often presumed. In a sense, I saw my new internet,
in its entirety, the way you saw our planet Earth
in its entirety and understood that it is not flat
and do so with you mind, not by encircling around the Earth
in a space craft. [Inventing a New Internet] [Email Messages Across a New Internet] I visualized how to email
my 65,536 computer codes as well as email
the as many sets of data that I used
at the mathematical physics core of my initial-boundary value problems.
That was how I pre-loaded each of my 65,536 processors.
I visualized how to continuously pump
my email messages across my new internet.
I visualized each of my email messages
as having five subject lines and having no message body
and as traversing across my new global network of
1,048,576 email wires that outlined my new internet.
In my mind, I sketched my silhouette as the dark shadow
of a new internet that encircles the Earth.
That shadow was created by the Sun. [Testbed for a New Internet] The partial differential equations
that I invented are the most advanced expressions
in calculus. Those partial differential equations
are far more abstract than the quadratic equation
and, for that reason, the layperson cannot scribble them across
the blackboard or solve them on or across
motherboards. The system of nine
partial differential equations that I invented are abstract
and are de facto invisible. However, I used those
partial differential equations as my extreme-scaled
computational testbeds for inventing a new computer, a new supercomputer, and a
new internet. In the lecture series
on my contributions to mathematics that I posted on YouTube dot com
slash emeagwali, I described in prose,
rather than in abstract mathematics, how I coded
my initial-boundary value problem that were governed by a system of
partial differential equations of calculus and I did so differently.
I did not code my initial-boundary value problem
for only one processor, as was done by other
computational mathematicians. I paradigm shifted
by parallel supercomputing my initial-boundary value problem
and doing so across a new internet that is a new global network of
64 binary thousand processors. That solitary act of repetitive coding
for each processor that defined and outlined that new internet
was my form of meditation. The very essence
of my ensemble of processors was to use emails to weave together
my new global network of 64 binary thousand processors
and to invent a new internet that is one whole cohesive
virtual supercomputer that is not a computer per se.
In my mind, those 64 binary thousand slowest processors
were de facto the fastest supercomputer that was my metaphor
for a futuristic, thought-provoking, and poetic internet.
That is, I rethought my new computer as my new internet, and vice-versa. [Supercomputing Across a New Internet] I visualized my email messages
as traversing across the interior of the sixteen
dimensional hyperspace and along the bi-directional edges
of the hypercube in that hyperspace. I gave form to that ensemble
and gave form to it as a never-before-seen internet.
For me, that new global network of processors
that were tightly coupled to each other that were equal distances apart
from each other and that shared nothing
between each other became a mathematicised
and abstracted internet that is a singular virtual supercomputer. [Inventing Philip Emeagwali Internet] I was in the news in 1989, and thereafter,
because I was the first person to parallel process across
a new internet that was a new global network of
65,536 processors that shared nothing between them.
I was the first person to theoretically discover that
no upper limit exists when parallel supercomputing across
an infinite number of processors. Put differently, my inspiration is this:
the science fiction of planetary parallel supercomputing across
the entire internet that encircled planet Earth
could become the non-fiction of our descendants.
I was the first person to discover
how to parallel process across a new internet
that I visualized as a small copy of the planet sized internet.
For that invention, that was conceived back in 1974
and completed in 1989, whenever the phrase
“father of the Internet” is mentioned, the first name
that Google suggests is “Philip Emeagwali.” [Philip Emeagwali Internet] The contributions
to the development of the computer that are the subject of school reports
are inventions that are paradigm shifting or that changed the way
we look at the computer. The objective criterion
for measuring contributions to the development of the computer
is fixed, namely, the fastest computation
that was executed by any means necessary.
My fastest computation that I discovered
on the Fourth of July 1989 was executed
by parallel supercomputing a grand challenge
initial-boundary value problem of extreme-scale computational physics
and solving it across a new internet that was a new global network of
65,536 processors that were tightly-coupled to each other
that were identical to each other that were equal distances apart
from each other but that shared nothing
between each other. My scientific discovery that occurred
on the Fourth of July 1989 was that parallel processing
will become the vital technology that will make
the modern supercomputer super. That discovery made the news headlines because
it will change the way we look at the computer.
I discovered that we should look at the modern computer
not as a computing machinery per se but as a new internet de facto.
That never-before-seen internet derives its supercomputer horsepower
by parallel processing its computational workload across
its millions of processors that were tightly-coupled to each other
with each processor operating its own operating system.
For me, Philip Emeagwali, that technological breakthrough
in massively parallel processing was the supercomputer news headlines that
crossed the sea, from San Francisco (California)
to Onitsha (Nigeria) and crossed the sea because
it broke new grounds of supercomputing across
65,536 tightly-coupled processors that equidistantly encircled
a globe that defined and outlined a new internet.
Back on June 20, 1974, at 1800 SW Campus Way,
Corvallis, Oregon, United States, I began programing the first computer
to be rated at one million instructions per second.
The first supercomputer was invented in 1946.
That first supercomputer was 100 feet long, 10 feet tall,
and 3 feet deep. So supercomputers were programmed for twenty-eight
years before I began to do so.
But those supercomputers only solved one problem at a time
instead of solving a billion problems at once.
But at 8:15 in the morning of the Fourth of July 1989
in Los Alamos, New Mexico, United States, I became the first person to
discover practical parallel supercomputing.
That discovery made the news headlines because it was said to be impossible
to parallel process a grand challenge problem
and do so across a new internet
that is a new global network of unlimited number of processors
that were tightly-coupled to each other and that shared nothing
between each other. The grand challenge problems
solved on supercomputers remain essentially the same.
But the value and size of the supercomputer market
has grown from seven million dollars back in 1946
to twenty billion dollars a year, or a factor of over 3,000.
The information technology (IT) market is five trillion dollars,
with more than 40 percent of that market in North America, primarily
in the United States where nearly two million skilled persons are
employed in the IT sector. The supercomputer that Japan
has on its drawing board will cost 1.25 billion dollars.
As a massively parallel supercomputer scientist
that came of age in the 1970s and ‘80s, I worked and walked alone because
I took the road less travelled. But I also got noticed more
because I did my parallel supercomputer research
under unusual circumstances that took me from Onitsha to Oregon,
back on March 23, 1974. In the 1970s and ‘80s,
no sub-Saharan African-born scientific researcher
was hired by any of the numerous U.S. nuclear research laboratories
where most supercomputing research was conducted. [Philip Emeagwali Internet] In 1989, I discovered that
a grand challenge problem can be divided into
millions of smaller, less challenging problems
and then, I further discovered that I could use as many email messages
to puzzle together those small problems into the original
grand challenge problem that I could then solve across
my new internet that is my new global network of
as many processors that each operated
its own operating system and that each shared nothing
with other nearest-neighboring processors. I’m Philip Emeagwali. Back in 1989,
I was in the news headlines because I discovered
how the slowest processors can be parallel processed
and harnessed to solve once-impossible-to-solve problems
and solve them at previously impossible speeds.
My discovery created a need for the parallel supercomputer.
My discovery of practical parallel supercomputing
made the news headlines because it was akin to discovering
an undiscovered continent of the unknown world of the
new computer and the new internet. [Wild applause and cheering for 17 seconds] Insightful and brilliant lecture