|
It is practically impossible, nowadays, to do science without
computers. Scientists are facing increasingly
complicated problems which require much more than a blackboard!
Often, a single computer, a cluster of standard computers or even a special-purpose
supercomputer, is not enough for the calculations scientists really
want to do. That's the way scientists are - always pushing the limits.
Of course, computers are improving incredibly fast. Still, they do not
keep up with what scientists demand of them.
As a result, scientists are often faced with situations
where they "hit the wall", and which make it very
difficult, very expensive, and sometimes downright impossible
to achieve certain scientific goals with current computer technology.
So some scientists started dreaming.
They dreamt of a way to surmount these obstacles. They dreamt
of having nearly infinite storage space so they would never have to worry
where to put the data. They dreamt of having nearly infinite computing
power available for their institution, whenever they need it. They
dreamt of being able to collaborate with distant colleagues easily
and efficiently, safely sharing with them resources, data, procedures
and results.
And, being always worried about their research grants, they
dreamt of doing all this very cheaply - maybe even for free! (Dreaming
costs nothing.)
  
Increasingly complicated problems?
| Ten
years ago, biologists were happy if they could
simulate a single small molecule on a computer, now they want to simulate
thousands of molecular drug candidates to see how they would interact
with specific proteins. |
Earth scientists
keep track of the level of atmospheric ozone with satellite observations.
For this task alone, they download, from space to ground, about
100 Gigabytes of raw images per day (the equivalent of about 150
CDs).
|
To
explore fundamental particles and the forces between them, High
Energy Physics will soon produce about 10 Petabytes of data per
year (about 20 million CDs). This data will record the result of collisions
of extremely energetic fundamental particles. Thousands of physicists
in dozens of universities around the world will want to analyse this
data!
|
Unlocking
the secrets of the human genome would be impossible
without the computerized analysis of massive amounts of data, including
the sequence of the three billion chemical units that comprise our
DNA, which is the genetic blueprint of our species. |
Typically, scientists hit the
wall when they are faced with situations where:
- The amount of data they need is huge and the data is stored in different
institutions. This could be the case, for example, of satellite images
of the Earth. It might take ages to copy the data one needs to one central
computer in order to analyse it. So ideally the scientist wants to do
the computation where the data is.
- The amount of similar calculations the scientist has to do is huge.
This could be the case, for example, when simulating the effect of thousands
of potential drug molecules on a protein related to some disease. This
would take ages on one computer, or even a cluster of computers.
- A scientific team with members around the globe wants to share large
amounts of data and do complex analysis of the data rapidly online together,
while discussing the results in a video conference.
|
