6 The Discovery of the Higgs Boson at the LHC
279
During the installation, the experiments made extensive use of the constant flow
of cosmic rays impinging on Earth providing a reasonable flux of muons even at
a depth of 100 m underground. Typically a few hundred per second traverse the
detectors. These muons were used to check the whole chain from sub-detector
hardware to analysis programs of the experiments, and to align the detector elements
and calibrate their response prior to the proton-proton collisions. In particular, after
the LHC incident on 19th September 2008 the experiments used the 15 months
LHC down time, before the first collisions on 23rd November 2009, to run the
full detectors in very extensive cosmic-ray campaigns, collecting many hundreds
of millions of muon events. These runs allowed both ATLAS and CMS to be ready
for physics operation, with pre-calibrated and pre-aligned detectors, by the time of
the first pp collisions.
6.5 Experiment Software and LHC Worldwide Computing
Grid
The experiment collaborations themselves develop the software that enables reconstruction, from raw data, of analyzable objects such as electrons, photons, jets, b
jets, muons, and other charged tracks, and their energies or momenta. Algorithms
have to be run to calibrate the energy deposits; align the hits from charged particles;
and correct for changes in detector response arising from irradiation, variation
in environmental parameters such as temperature, or changes in the position of
detecting elements. The software packages must also simulate the response of
the detectors to the passage of particles generated in simulated events occurring
in bunch crossings that contain interesting physics processes, as well as simple
backgrounds. These include processes such as the production of W or Z bosons,
QCD jets, or Higgs bosons and their decays. Such simulations helped prepare, prior
to the first collisions, the experiments’ end-to-end processing and analysis chains,
which were crucial for the rapid delivery of physics results of outstanding quality
and quantity soon after the first collisions.
The LHC computing system, termed the LHC Worldwide Computing Grid
(WLCG) [1], was conceived to make effective use of distributed resources, work
on a large scale, and enable all the experiments’ scientists, wherever they were
based, to have access to LHC data, and without regard to the extent of the resources
they themselves could afford. The WLCG provided the backbone for the analysis
capabilities of the experiments. The global WLCG has continued to grow, now
encompassing around 170 computing centers in 42 countries, with an infrastructure
that provides access to some 600,000 computing cores, around 500 PB of storage
(50% on disk and 50% on tape), and a network that frequently runs at 100 Gb s −1
between larger sites and at 10 Gb s −1 between smaller sites. The security of access
have been instrumental for building a truly federated computing infrastructure for
science. Although the individual computing tasks described above were already
279
During the installation, the experiments made extensive use of the constant flow
of cosmic rays impinging on Earth providing a reasonable flux of muons even at
a depth of 100 m underground. Typically a few hundred per second traverse the
detectors. These muons were used to check the whole chain from sub-detector
hardware to analysis programs of the experiments, and to align the detector elements
and calibrate their response prior to the proton-proton collisions. In particular, after
the LHC incident on 19th September 2008 the experiments used the 15 months
LHC down time, before the first collisions on 23rd November 2009, to run the
full detectors in very extensive cosmic-ray campaigns, collecting many hundreds
of millions of muon events. These runs allowed both ATLAS and CMS to be ready
for physics operation, with pre-calibrated and pre-aligned detectors, by the time of
the first pp collisions.
6.5 Experiment Software and LHC Worldwide Computing
Grid
The experiment collaborations themselves develop the software that enables reconstruction, from raw data, of analyzable objects such as electrons, photons, jets, b
jets, muons, and other charged tracks, and their energies or momenta. Algorithms
have to be run to calibrate the energy deposits; align the hits from charged particles;
and correct for changes in detector response arising from irradiation, variation
in environmental parameters such as temperature, or changes in the position of
detecting elements. The software packages must also simulate the response of
the detectors to the passage of particles generated in simulated events occurring
in bunch crossings that contain interesting physics processes, as well as simple
backgrounds. These include processes such as the production of W or Z bosons,
QCD jets, or Higgs bosons and their decays. Such simulations helped prepare, prior
to the first collisions, the experiments’ end-to-end processing and analysis chains,
which were crucial for the rapid delivery of physics results of outstanding quality
and quantity soon after the first collisions.
The LHC computing system, termed the LHC Worldwide Computing Grid
(WLCG) [1], was conceived to make effective use of distributed resources, work
on a large scale, and enable all the experiments’ scientists, wherever they were
based, to have access to LHC data, and without regard to the extent of the resources
they themselves could afford. The WLCG provided the backbone for the analysis
capabilities of the experiments. The global WLCG has continued to grow, now
encompassing around 170 computing centers in 42 countries, with an infrastructure
that provides access to some 600,000 computing cores, around 500 PB of storage
(50% on disk and 50% on tape), and a network that frequently runs at 100 Gb s −1
between larger sites and at 10 Gb s −1 between smaller sites. The security of access
have been instrumental for building a truly federated computing infrastructure for
science. Although the individual computing tasks described above were already