How I Invented My New Internet | A Father of the Internet | Philip Emeagwali Internet

How I Invented My New Internet | A Father of the Internet | Philip Emeagwali Internet


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] [Struggles Against Dogma] Back in the 1970s and ‘80s,
almost every vector processing supercomputer scientist
believed that parallel processing is a huge waste of everybody’s time.
So, I was executing my massively parallel processing experiments
and executing them against the orders
of the leaders of thought in the world of computing
—such as the Steve Jobs of personal computing—
and against the opinions of the leaders of thought
in supercomputing, such as Gene Amdahl and Seymour Cray.
In the 1970s and ‘80s, the terra incognita
that was the emerging field of massively parallel processing supercomputing
was as empty as a ghost town that had only one permanent resident.
I—Philip Emeagwali— was that permanent resident
of the farthest frontier of supercomputing called massively parallel processing.
In the 1970s and ‘80s, if you could find your way
to any massively parallel processing supercomputer, its administrator will deem you worthy
and grant you a supercomputer account to become its lone wolf programmer.
Because the internet of the early 1980s was then uncrowded,
I had an unusual email address from the early 1980s.
That email address had no dot com suffix.
In the mid-1980s, I had the email address spelled
emeagwal @ think dot com. Emeagwal was spelled like my last name
without the last letter “i.” Think dot com
was the second registered dot com suffix. [Visualizing a Small Copy of the Internet] That global network of
64 binary thousand commodity-off-the-shelf processors
that I experimentally discovered that it could be programmed
to solve the toughest problems in computational physics
was a new internet. That new internet
was a small copy of a never-before-understood Internet,
that had only 65,536 processors around a globe
instead of billions of computers around a globe.
I visualized each of my two-to-power sixteen
commodity processors as equal distances apart
from each other and around a globe
in a sixteen-dimensional hyperspace. And I visualized my ensemble
of processors as evenly distributed across
the hypersurface of a hypersphere in a sixteen-dimensional universe.
I visualized my ensemble of processors
as outlining a new internet that I visualized
in my sixteen-dimensional universe. [David Versus Goliath] I—Philip Emeagwali—was the David
from the world of the massively parallel processing supercomputer
that was ridiculed and mocked for challenging the Goliath
—named Seymour Cray—who designed seven in ten supercomputers
in the world of the vector processing supercomputer of the 1970s and ‘80s.
I visualized my massively parallel processing supercomputer
as my sling shot that is a small copy of the Internet
that can shoot 65,536 small pebbles
from its as many processors. Those pebbles were my metaphors
for the as many initial-boundary value problems
of modern calculus and computational physics.
I visualized shooting all the 65,536 small pebbles
at once. I can only record the fastest computations
and record them across 65,536
processors and record them
by throwing all my rocks at once,
instead of throwing them one at a time.
In the 1970s and ‘80s, I was the David of supercomputing.
I was ridiculed and caricatured by well-regarded supercomputer scientists.
I was called a “lunatic” and dismissed from research teams
that believed that all supercomputers
must do only one thing at a time. Seymour Cray
—the Goliath of supercomputing— believed that all supercomputers
should compute only one thing at a time. Seymour Cray
was armed with one big sword. Seymour Cray’s sword
was my metaphor for his vector processing supercomputer.
Seymour Cray’s most famous quote is this: [quote]
“If you were plowing a field, which would you rather use?
Two strong oxen or 1024 chickens?”
[unquote] As reported in the June 20, 1990 issue
of The Wall Street Journal, I—Philip Emeagwali—
experimentally discovered that 65,536 chickens
that learned to work together, or work in parallel,
can plow more field than the strongest ox that works alone. [How I Invented a New Internet] I’m Philip Emeagwali.
I’m the subject of school reports because I invented
a new supercomputer that was the precursor
to the modern supercomputer. I invented a new supercomputer
that is a small copy of a new internet.
The new internet that I invented is defined and outlined by an ensemble
of 65,536 commodity-off-the-shelf processors
that are identical and that are equal distances
apart. That
new internet is complex, abstract, and a mystery.
The 65,536 processors of my new internet were married together
by 1,048,576 bi-directional email wires and married together
as a new supercomputer that computed cohesively
and did so as one new integrated supercomputer and communicated seamlessly
as one new internet. My 64 binary thousand processors
that outlined my new internet communicated via emails
and did so with a complexity that I cannot completely describe
in words alone. Nor can I completely describe
my processor-to-processor email exchanges and completely describe them as equations
on a blackboard alone or completely describe them as algorithms
on a motherboard alone. I began supercomputing at age 19
on June 20, 1974 in Corvallis, Oregon, United States.
I was the lone wolf and the only full time programmer
of the fastest supercomputer of the 1980s.
Today, the fastest supercomputer costs the budget of a small nation.
The fastest supercomputer is programmed by thousands
of supercomputer scientists. The fastest supercomputer
occupies the space of a soccer field. The Holy Grail of the fastest possible supercomputer
is to marry together all the processors in the world
and marry them to all the computers in the world
and marry them to all the supercomputers
in the world and marry processors and computers
and supercomputers together and as a never-before-seen internet
that will become a never-before-seen planetary-sized supercomputer
that will turn our science fiction to our descendant’s non-fiction. I’m Philip Emeagwali. [How I Named a New Internet] I’m Philip Emeagwali.
I’m the massively parallel processing supercomputer scientist
that conducted research alone and conducted it from the age of nineteen
in Corvallis, Oregon, United States to the age of 35
in Los Alamos, New Mexico, United States. To the supercomputer scientist,
Los Alamos, New Mexico is the capital of supercomputing.
Prior to my experimental discovery of the massively parallel processing supercomputer
that occurred on the Fourth of July 1989, it was said that
parallel processing is a beautiful theory
that lacked experimental confirmation. Prior to the Fourth of July 1989,
I was the unknown supercomputer scientist who told his massively parallel processing
supercomputer story alone and told it to no supercomputer scientist
in particular. In fact, my 1,057-page report
that I distributed to vector processing
supercomputer scientists of the 1980s
and that described my new supercomputer was, at first, thrown into the trash.
After my experimental discovery of the massively parallel processing supercomputer
that occurred on the Fourth of July 1989, I became a known supercomputer scientist
and those that threw my 1,057-page report into the trash
wanted to become my new best friend and clamored
to retell the story of how I experimentally discovered
the massively parallel processing supercomputer that is a new internet.
As I became more known, I discovered that
many insidious voices were clamoring to retell my story
and to retell it in their visions,
rather than in my original vision. I discovered that
their thousand secondary voices can drown my primary voice.
I discovered that the story of my new supercomputer
that is not a computer per se but that is a new internet de facto
was reduced to a cacophony of secondary voices.
I want to redeem my story and reclaim my voice
and make my voice the loudest voice
in the world of the modern supercomputer and make my voice
to be the most continuously heard voice in the history of the Internet. [The Magic Zone: Naming My New Internet] Each of my processor—within my ensemble
of 65,536 processors— had its unique name
that’s also its unique email address that’s sixteen bits long.
I used a binary reflected code to generate
my 64 binary thousand unique names that were each
a unique string of sixteen zeroes and ones.
With the binary reflected internet naming scheme
that I used, if two email addresses
differed by only one bit, then the processors
that corresponded to those two email addresses
differed by only one bit. And those two processors
were directly connected. That connection
allows nearest-neighbor email communications that maximizes
the speed I could attain while executing
my floating-point arithmetical operations. My sixty-four binary thousand emails
travelled across 0ne binary million,
or one million, forty-eight thousand five hundred and seventy-six [1,048,576],
bi-directional email wires. [Wild applause and cheering for 17 seconds] Insightful and brilliant lecture

1 Comment

  • Philip Emeagwali says:

    I'm Philip Emeagwali. Back in the 1980s, almost every vector processing supercomputer scientist believed that parallel processing is a huge waste of everybody’s time. So, I was executing my massively parallel processing experiments and executing them against the orders of the leaders of thought in the world of computing —such as the Steve Jobs of personal computing—and against the opinions of the leaders of thought in supercomputing, such as Gene Amdahl and Seymour Cray. In the 1970s and ‘80s, the terra incognita that was the emerging field of massively parallel processing supercomputing was as empty as a ghost town that had only one permanent resident. I—Philip Emeagwali—was that permanent resident of the farthest frontier of supercomputing
    called massively parallel processing.

    In the 1970s and ‘80s,

    if you could find your way

    to any massively parallel processing supercomputer,

    its administrator will deem you worthy

    and grant you a supercomputer account

    to become its lone wolf programmer.

    Because the internet of the early 1980s

    was then uncrowded,

    I had an unusual email address

    from the early 1980s.

    That email address

    had no dot com suffix.

    In the mid-1980s,

    I had the email address spelled

    emeagwal @ think dot com.

    Emeagwal was spelled like my last name

    without the last letter “i.”

    Think dot com

    was the second registered

    dot com suffix.

    Visualizing a Small Copy of the Internet

    That global network of

    64 binary thousand

    commodity-off-the-shelf processors

    that I experimentally discovered

    that it could be programmed

    to solve the toughest problems

    in computational physics

    was a new internet.

    That new internet

    was a small copy

    of a never-before-understood Internet,

    that had only 65,536 processors

    around a globe

    instead of billions of computers

    around a globe.

    I visualized each of my

    two-to-power sixteen

    commodity processors

    as equal distances apart

    from each other

    and around a globe

    in a sixteen-dimensional hyperspace.

    And I visualized my ensemble

    of processors

    as evenly distributed across

    the hypersurface of a hypersphere

    in a sixteen-dimensional universe.

    I visualized my ensemble

    of processors

    as outlining a new internet

    that I visualized

    in my sixteen-dimensional universe.

    David Versus Goliath

    I—Philip Emeagwali—was the David

    from the world of the massively

    parallel processing supercomputer

    that was ridiculed and mocked

    for challenging the Goliath

    —named Seymour Cray—who designed

    seven in ten supercomputers

    in the world of the vector processing supercomputer

    of the 1970s and ‘80s.

    I visualized my massively

    parallel processing supercomputer

    as my sling shot

    that is a small copy of the Internet

    that can shoot

    65,536 small pebbles

    from its as many processors.

    Those pebbles were my metaphors

    for the as many

    initial-boundary value problems

    of modern calculus

    and computational physics.

    I visualized shooting

    all the 65,536 small pebbles

    at once.

    I can only record the fastest computations

    and record them across

    65,536

    processors

    and record them

    by throwing all my rocks

    at once,

    instead of throwing them

    one at a time.

    In the 1970s and ‘80s,

    I was the David of supercomputing.

    I was ridiculed and caricatured

    by well-regarded supercomputer scientists.

    I was called a “lunatic”

    and dismissed from research teams

    that believed that

    all supercomputers

    must do only one thing at a time.

    Seymour Cray

    —the Goliath of supercomputing—

    believed that all supercomputers

    should compute only one thing at a time.

    Seymour Cray

    was armed with one big sword.

    Seymour Cray’s sword

    was my metaphor

    for his vector processing supercomputer.

    Seymour Cray’s most famous quote is this:

    [quote]

    “If you were plowing a field,

    which would you rather use?

    Two strong oxen

    or 1024 chickens?”

    [unquote]

    As reported in the June 20, 1990 issue

    of The Wall Street Journal,

    I—Philip Emeagwali—

    experimentally discovered that

    65,536 chickens

    that learned to work together,

    or work in parallel,

    can plow more field

    than the strongest ox that works alone.

    How I Invented a New Internet

    I’m Philip Emeagwali.

    I’m the subject of school reports

    because I invented

    a new supercomputer

    that was the precursor

    to the modern supercomputer.

    I invented a new supercomputer

    that is a small copy

    of a new internet.

    The new internet that I invented

    is defined and outlined by an ensemble

    of 65,536

    commodity-off-the-shelf processors

    that are identical

    and that are equal distances

    apart.

    That new internet

    is complex, abstract, and a mystery.

    The 65,536 processors of my new internet

    were married together

    by 1,048,576 bi-directional email wires

    and married together

    as a new supercomputer

    that computed cohesively

    and did so as one new integrated supercomputer

    and communicated seamlessly

    as one new internet.

    My 64 binary thousand processors

    that outlined my new internet

    communicated via emails

    and did so with a complexity

    that I cannot completely describe

    in words alone.

    Nor can I completely describe

    my processor-to-processor email exchanges

    and completely describe them as equations

    on a blackboard alone

    or completely describe them as algorithms

    on a motherboard alone.

    I began supercomputing at age 19

    on June 20, 1974

    in Corvallis, Oregon, United States.

    I was the lone wolf

    and the only full time programmer

    of the fastest supercomputer

    of the 1980s.

    Today, the fastest supercomputer

    costs the budget of a small nation.

    The fastest supercomputer

    is programmed by thousands

    of supercomputer scientists.

    The fastest supercomputer

    occupies the space of a soccer field.

    The Holy Grail of the fastest possible supercomputer

    is to marry together

    all the processors in the world

    and marry them

    to all the computers in the world

    and marry them

    to all the supercomputers

    in the world

    and marry processors and computers

    and supercomputers together

    and as a never-before-seen internet

    that will become a never-before-seen planetary-sized supercomputer

    that will turn our science fiction

    to our descendant’s non-fiction.

    I’m Philip Emeagwali.

    How I Named a New Internet

    I’m Philip Emeagwali.

    I’m the massively parallel processing

    supercomputer scientist

    that conducted research alone

    and conducted it from the age of nineteen

    in Corvallis, Oregon, United States

    to the age of 35

    in Los Alamos, New Mexico, United States.

    To the supercomputer scientist,

    Los Alamos, New Mexico

    is the capital of supercomputing.

    Prior to my experimental discovery

    of the massively parallel processing supercomputer

    that occurred on the Fourth of July 1989,

    it was said that

    parallel processing

    is a beautiful theory

    that lacked experimental confirmation.

    Prior to the Fourth of July 1989,

    I was the unknown supercomputer scientist

    who told his massively parallel processing supercomputer story alone

    and told it to no supercomputer scientist

    in particular.

    In fact, my 1,057-page report

    that I distributed

    to vector processing

    supercomputer scientists

    of the 1980s

    and that described my new supercomputer

    was, at first, thrown into the trash.

    After my experimental discovery

    of the massively parallel processing supercomputer

    that occurred on the Fourth of July 1989,

    I became a known supercomputer scientist

    and those that threw my 1,057-page report into the trash

    wanted to become my new best friend

    and clamored

    to retell the story

    of how I experimentally discovered

    the massively parallel processing supercomputer

    that is a new internet.

    As I became more known,

    I discovered that

    many insidious voices

    were clamoring to retell my story

    and to retell it

    in their visions,

    rather than in my original vision.

    I discovered that

    their thousand secondary voices

    can drown my primary voice.

    I discovered that

    the story of my new supercomputer

    that is not a computer per se

    but that is a new internet de facto

    was reduced to a cacophony

    of secondary voices.

    I want to redeem my story

    and reclaim my voice

    and make my voice

    the loudest voice

    in the world of the modern supercomputer

    and make my voice

    to be the most continuously heard voice

    in the history of the Internet.

    The Magic Zone: Naming My New Internet

    Each of my processor—within my ensemble of 65,536 processors—

    had its unique name

    that’s also its unique email address

    that’s sixteen bits long.

    I used a binary reflected code

    to generate

    my 64 binary thousand unique names

    that were each

    a unique string of

    sixteen zeroes and ones.

    With the binary reflected

    internet naming scheme

    that I used,

    if two email addresses

    differed by only one bit,

    then the processors

    that corresponded

    to those two email addresses

    differed by only one bit.

    And those two processors

    were directly connected.

    That connection

    allows nearest-neighbor email communications that maximizes

    the speed I could attain

    while executing

    my floating-point arithmetical operations.

    My sixty-four binary thousand emails

    travelled across

    0ne binary million,

    or one million, forty-eight thousand

    five hundred and seventy-six [1,048,576],

    bi-directional email wires.

Leave a Reply

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