Information Technology Reference
In-Depth Information
5.6.2 CERN (Business Case 4)
The example of this section corresponds to the fourth category of table 5.1 and
table 5.2. A representative real-life scenario of this specific business case is CERN,
the European Organization for Nuclear Research, one of the world's largest and
most respected centers for scientific research. Its subject is fundamental physics,
namely finding out what the Universe is made of and how it works. The world's
largest and most complex scientific instruments are used in CERN to study the basic
constituents of matter, i.e. the fundamental particles. By studying what happens
when these particles collide, physicists learn about the laws of Nature. One of the
largest experiments that is currently in progress is the Large Hardon Collider project
(LHC). The main purpose of LHC is to discover more about how the universe began
and what it's made of. This will be achieved by colliding beams of protons and ions
at a velocity approaching the speed of light. Those records generated by this experi-
ment are predicted to occupy 15 Petabytes of memory every year, an enormous
amount of data that cannot be accommodated either by a single IT infrastructure
nor by a supercomputer. However, thankfully, Grid technology provides a solution
to effectively store and process this huge amount of data.
CERN leads a major Grid project, the LHC Computing Grid, which is dedicated
to providing the processing power and storage capacity necessary for the LHC
.
Grid
was adopted because of the benefits Grid provides, such as the much lower cost,
the flexibility and the ease of upgrades, compared to a single large and complex
machine. In order for the network of computers to be able to store and analyze data
for every experiment conducted at the LHC, a special middleware for Grid archi-
tecture was developed.
The structure of the system is organized into three tiers. The first is CERN's
computing system, which is dedicated to process the information at the beginning
and divide it into chunks for the other tiers. There are twelve second-tier sites that
are located in several countries whose purpose is to accept data from CERN over
dedicated computer connections. When the LCG gets up to full capacity, it will be
made up of around 200,000 processors, mostly located in 11 academic computing
clusters around the world, as shown in table 5.4, that will let around 7,000 scien-
tists conduct experiments related to the collider, submitting their calculations to the
LCG, which will farm them out around the network according to the supply and
demand for resources (Cern 2006, Johnson 2008, Ranger 2005).
