 |
Newsletter |
|
|
|
|
|
Contacts |
IFCA-CSIC
Dr. Celso Martinez Rivero
CYFRONET
Prof. M. Turala
|
|
 |
|
|
|
|
|
The
next generation of high-energy physics experiments will require
unprecedented computing resources for data analysis. The
CrossGrid aim is to
develop end-user applications for physics analysis
running in a distributed
mode, in a Grid-aware environment, using large
distributed databases.
Distributed data mining techniques will reduce the
waiting time in these
analyses from hours to minutes, allowing tests of
powerful strategies on
large databases. The use of distributed computing
resources for real-time
filtering is also being studied. These applications will
be employed in the
high-energy physics field, the main focus being the
future LHC experiments
(ALICE,
ATLAS, CMS and LHCb).
|
|
|
|
|
|
|
 |
High Energy Physics
Understanding the building blocks of the
Universe
|
Complex
particle accelerators recreate the conditions at
the first moments of the Universe,
in order to study the forces between elementary
particles.
The next Large
Hadron Collider (LHC) at CERN will accelerate
protons to an energy enough to produce a
particle hundreds of times heavier: the Higgs
Boson, the last piece in the Standard Model, key
for understanding the origin of mass.
All collisions
will be recorded by sophisticated detectors, the
corresponding information being
stored in distributed databases with a volume of
millions of gigabytes.
But only few of those complex collisions
will produce a Higgs Boson…
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
On-line filtering
techniques as well as sophisticated
mathematical algorithms,
such as
neural networks, will be used to select
those events and analyze them by
physicists working in research centres
across the
whole
world. They will share
their computing resources in a large
GRID framework.
Distributed data-mining techniques
developed in the CROSSGRID project will
reduce the waiting
time in these analyses from hours to
minutes, allowing the test of powerful
strategies over larger datasets
An intuitive Web
Portal will act as
a
user interface to
the
GRID resources,
hiding all the complex “behind-the-scene”
processes:
-
generic
access to large
distributed databases
-
load
balance across the
computing nodes, prioritizing
interactive use
-
monitoring
and final result display
|
|
Testing Distributed
Computing Techniques in the GRID
framework
|
A multidisciplinary approach
The
distributed computing techniques for
database access and data-mining are based on
open source software and standards, such as
XML, MPI and the Globus toolkit; these will
be applied within CROSSGRID to the
Meteorology and Environmental fields.
Partnership and
the
CROSSGRID testbed
The
Institute of Physics of Cantabria (IFCA,
CSIC) in Santander (SPAIN) is
leading the development of this
application, in collaboration with
IFIC (Valencia), INP (Krakow), UAB
(Barcelona), FZK (Karlsruhe) and INS
(Warsaw).
The
CROSSGRID testbed, joining computing
resources in 15 sites across Europe,
through the Géant network, is used
for testing and deployment of this
software in the Grid framework.
CROSSGRID,
an
EC-funded
R&D project, involves 21 partners
and is coordinated by CYFRONET, the
Academic Computing Center in Krakow,
Poland.
The dissemination and
exploitation of the results of the
CROSSGRID project is coordinated by
Algosystems S. A., an IT company
based in Athens, Greece.
|
|
|
|
|
|
|
|
|
|
|
| |
|
|
|
Experiment data-chain filtering structure
|
|
|
|
|
Particle
selection inside the LHC
collision
|
|
|
|
Partners & the CROSSGRID testbed
|
|
|
