To add comments or start new threads please go to the full version of: Japan to develop world's fastest supercomputer
PhysOrgForum Science, Physics and Technology Discussion Forums > PhysOrg WebLog > PhysOrg WebLog

bob
Japanese companies, universities and a government-affiliated research facility plan to jointly develop the world fastest supercomputer, capable of 10 quadrillion floating-point operations per second. NEC and Hitachi are expected to take part.

Work is due to begin in the year starting April 2006, with the aim of having the supercomputer operational by March 2012.

The next-generation supercomputer is expected to help analyze the structures of disease-causing proteins and it will also be used for nuclear fusion research.
ap2
That will be quite a breakthrough from current record of 135 trillion floating point operations per second
Hello
10 quadrillion = 1000 trillion

Almost a tenfold increase. Not bad!
hum sorry ?
i think that you have made a mistake : 10 q = 10000 t
SS
1 Trillion = 10^18
1 Quadrillion = 10^24
10 Qaudrillion = 10 x 10^6 Trillion
JavaTool
Actually, a quadrillion would be 10^15 and a trillion is 10^12. 10 quadrillion would therefore be 10^16. Come on, guys... rolleyes.gif
vhawk
That is what I like about these term ... i.e. Trillion , Quadrillion etc.

Common guys, his is 2005 - who is going to express that in some other non defined UOM. It is a 'zillion' times easier jut to say (something) x 10^(something) at least then any one can work out how many zero's to add .... laugh.gif
Guest
I predict that it will have 4 cyclnders, get better mileage, and be of better quality than the american designs.

I predict that the american companies will not change their ways, and continue to be arrogant and close minded.

I predict that american companies will ask the government to protect them.

Oh, wait, wrong subject, never mind.
nellatol
Knowing fully well that even a Mouse's Brain is far far superior even to modern day's superdupercomputers it is a big waste of energy.

It may be true that the Brain is highly dynamic. But it also is very very true that the Brain can recognize a person or remember a thing even after decades!

So, 75% of the struggled should be strictly focussed on finding a niche in say mouse, rat or bat Brain and use them as our day-to-day SuperUltra_Computers!

A live Animal with its Magnificient Brain sitting in a cage is waiting to be USED!

Find a way! Find a niche! in IT to find all the damn answers for HURRICANES, EARTHQUAKES, DRUG_DISCOVERIES LIFESOULDEATHGOD AND what not!?
SS
Some info about the Quadrillion
http://www.dagmar-mueller.de/wdz/Zahlen/Za...e/vorsatze.html
quadrillius
Hmm, I'm from europe and I dunno about anyone else but I can't persuade people that a billion is a thousand million, let alone a quadrillion.

by my reckoning:

1000 thousand
1000000 million
1000000000 billion
1000000000000 trillion
1000000000000000 quadrillion

therefore: quadrillion = 10^15

for those who misunderstood the first comment

US billion (now widly accepted) = 10^9
US Milliard (I belive now known as a trillion) = 10^12

EU billion (used by people who won't get with the programme) = US Milliard
nellatol
MICE ARE BETTER ROBOTS.

1 Mouse = 1 Quadrillion Robots.

Mice have better eyes.
Have better ears.
As long as they are fed & aren't dead.
As long as they wouldn't start wagging their tails.

Little related power usage; Little space occupance.

You switch to bats. You get additional functionalities. FLYING & Radar?

How do you approach the problem. Let WIMAX come in full force.

Make a wireless contact with their Brains (minute implants in their brains, so to say) and start BUGGING THOSE Cerebromputers!

I am sure they don't mind. Atleast they do not have no choice.

As I read a comment earlier elsewhere, animal networking using 3 to 4 piece cellphones in a pack of large dogs should not be a bad idea. They can initiate call to their peers by bring their front paws (rigged with cellphone components) together!...at least they can be taught to by applying carrot & stick approach. Humans interacting with machines is less exciting relative to a pack of Dogs interacting among themselves!
George
It will be nice if the team is led remotely (through cellphone or color-coded lasers) by a human, though.

As for the pack of dogs, 1 dog should be a not necessarily ferocious mom-dog and the others child-dogs spread out (in other words, with no olfactory cues whatsoever).

The lead dog should get its remote command from the Human and it in its turn should commandeer its dog-forces to a different site each time through remote calls!

It will be a real nice demonstration keeping up with modern times!

In fact soon afterwards the lead dog should learn to play these games (again taught through carrot and stick approach) with its subordinates all day long... I mean without any human intervention.

Of course the key remains as suggested in the other's comment that the cell phone has to be divided into a few lighter components (all connected wirelessly amongthemselves)..one light piece near each ear..one light piece near the mouth and one light piece on each paw. The dogs SHOULD INITIATE the call by bringing the 2 front paws togehter. The cell phone may be made to start based on sound....but dogs may indulge in barks extraneous to the study. Hearing part should be on all the time probably (or not).
PhysOrg scientific forums are totally dedicated to science, physics, and technology. Besides topical forums such as nanotechnology, quantum physics, silicon and III-V technology, applied physics, materials, space and others, you can also join our news and publications discussions. We also provide an off-topic forum category. If you need specific help on a scientific problem or have a question related to physics or technology, visit the PhysOrg Forums. Here you’ll find experts from various fields online every day.
To quit out of "lo-fi" mode and return to the regular forums, please click here.
©PhysOrg.com - physics and technology news - Version for PDAs