ATLAS and CMS are particle physics experiments on the Large Hadron Collider at CERN, Geneva. These detectors are searching for new discoveries in the collisions of protons of extraordinarily high energy.

These experiments will learn about the basic forces that have shaped our Universe since the beginning of time and that will determine its fate.

The experiments are currently running a large scale distributed computing system to process the data acquired by the experiment. The ATLAS Distributed Computing environment (ADC) consists of several pieces: a Distributed Data Management component (DDM), a distributed production and analysis system (PANDA), and associated tools as well as the processing and analysis binaries. This flagship is part of a wider project within ATLAS, to research the applicability of cloud computing to ATLAS computing.

“CERN’s computing capacity needs to keep-up with the enormous amount of data coming from the Large Hadron Collider and we see Helix Nebula- the Science Cloud as a great way of working with industry to meet this challenge,” says Frédéric Hemmer, head of CERN’s IT department.

What we want to achieve.

• Evaluating the use of cloud technologies for LHC data processing,
• Transparent integration of cloud computing resources with ADC software and services,
• Implementation of the ATLAS cloud computing model within DDM, Panda, and related tools and services.

What we would like to know.

• What are the financial costs of network transfers of data into and out of the cloud resources, short and long-term data storage in the cloud, and CPU resources for running the various ATLAS use cases?
• What are appropriate Service Level Agreements and how can they be defined for broader use?
• What are the policy and legal constraints in moving scientific data across academic networks into commercial resources and back again?

How will the project assist the scientific community?

The ATLAS scientific community of ~3000 physicists will directly benefit from this work. In the short term the benefits of running the ATLAS software in a commercial cloud system will include the possibility to dynamically acquire additional resources when needed. In the longer term, use of commercial resources could become a real alternative to very large data centres owned and managed by the scientific community.