Loading...

Philip Emeagwali vs. Steve Jobs on Changing the Way We Look at Computers

32 views

Loading...

Loading...

Transcript

The interactive transcript could not be loaded.

Loading...

Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Jul 18, 2019

I’m Philip Emeagwali at http://emeagwali.com. Thirty-two [32] years after that negative article appeared in the Computer World magazine, Steve Jobs also questioned the usefulness of research in parallel processing. Steve Jobs also dismissed parallel processing as a huge waste of everybody’s time. On June 9, 2008, Steve Jobs told the opening session of Apple’s Worldwide Developers Conference in San Francisco, California that his research computer scientists at his Apple Corporation questioned the importance of research in parallel processing. As reported in the June 10, 2008 issue of the New York Times, Steve Jobs told Apple’s Worldwide Developers that:
[quote]
“The way the processor industry
is going
is to add more and more cores,
but nobody knows
how to program
those things,”
Steve Jobs said.
And Steve Jobs continued:
“I mean, two, yeah;
four, not really;
eight, forget it.”
[unquote]

5.6.2 Changing the Way We Think About Computers

July 4, 1989, the U.S. Independence Day,
was the day I changed the way
I look at the computer.
Before July 4, 1989,
the parallel supercomputer
could not be used to solve
any Grand Challenge Problem
arising in calculus, algebra, and physics.
On the Fourth of July 1989,
I became the first person
to achieve the theoretical speed up limit
for the parallel supercomputer
and to achieve a speed up
of a factor of 65,536.
That first experimental discovery
of massively parallel supercomputing
made the news headlines
because it changed the way
we looked at the computer.
In its first three decades,
parallel processing
was used to solve
embarrassingly parallel problems,
such as calculating the sum
of an array of numbers.
Or finding the maximum value
of an array of numbers.
Or multiplying two matrices.
Or finding the transpose of a matrix.
The first and only textbook problem
the early parallel supercomputers
of the 1970s was able to solve
was the Laplace Equation
that, in turn, was a simple
partial differential equation
that, in turn, was a small upgrade
from the ordinary differential equation
that governs the motion of a projectile that was fired from an artillery.
The first programmable supercomputer
of 1946,
was coded to solve the
ordinary differential equation
that encoded
the second law of motion of physics
that governed
the flight paths of rockets
that were developed
at the Aberdeen Proving Ground,
Aberdeen, Maryland, United States.
My first parallel processed supercomputer calculation
that was the world’s fastest
that occurred
on the Fourth of July 1989,
was coded to solve
the system of coupled, non-linear,
time-dependent, and three-dimensional
partial differential equations
of calculus
that also encoded
the second law of motion of physics
that governed
the [quote unquote] “flight paths”
of flowing fluids,
such as the flowing air and moisture
that flows above the surface of the Earth
and the flowing crude oil, natural gas,
and injected water
that flows one-mile deep
and below the surface of the Earth
and flows across a production oilfield
that is the size of a town.
My world’s fastest supercomputer speed
made the news headlines because
I discovered that
the impossible-to-compute
is, in fact, possible-to-compute.
In a metaphorical sense, Gene Amdahl
was arguing that it will be impossible
for nine women
to have one baby in one month,
instead of in nine months.
In my metaphorical sense,
I discovered that
nine women can have nine babies
in nine months and have them,
if and only if,
they conceived those nine babies
in parallel.
Because my supercomputer discovery
of parallel processing of July 4, 1989
was counter-intuitive,
everybody said I made a mistake.
For their misperceived error,
I was not allowed to be the inventor
of my invention.
I’m Philip Emeagwali.
I am the supercomputer scientist
that figured out
how to solve Grand Challenge Problems
and how to solve them
by parallel processing them
so that each processor solves
a different piece
of the Grand Challenge Problem
and all processors
solves them synchronously
and solves at the same time.
On the Fourth of July 1989
in Los Alamos, New Mexico,
United States,
I recorded an increase
in supercomputer speed
of a factor of 65,536.
My speed increase
made the news headlines
as a world record.
I recorded that increase across
as many processors
that were simultaneously solving
the Grand Challenge Problem
that I chopped up
into as many problems.
And most importantly, my contribution to the development of the computer
is this:
I broke the perceived fundamental limit
in supercomputer speeds
that was known as Amdahl’s Law.
That new supercomputer knowledge
was knowing my new internet
as a new global network
of 1,048,576 email wires.
I visualized
one binary million email wires
that tightly circumscribed
the hyper-surface of a hyper-globe
within a sixteen-dimensional hyperspace.

Philip Emeagwali 190610 2 6 of 6

Loading...


to add this to Watch Later

Add to

Loading playlists...