Running simulation ?
ATLAS at HOME tasks can be in principle of any type but, for the time being volunteers are running "Simulation" for two reasons : it is the part which can use a lot of CPU without much Input/Output traffic, and the work can be spread over the complete year... whereas ATLAS real data reconstruction campaigns and analysis are driven by the accelerator and conferences agenda, with very high stress peaks that you do not want to get into !
Our simulation needs are entirely driven by the LHC collider work plan and ATLAS hottest physics subjects. You will find the current WU (working units) description in the "Current tasks" menu on the right hand side of this page and, depending on the participants feedback and questions, key words definitions will be added in the "HEP glossary" (see top menu).
The application programs may change from time to time, if for example there is a change in the data format or improvement in the algorithms. It is very likely to be the case, for example, when ATLAS takes new data in 2015. This web site will explain why and the physics behind.
One needs a very large number of events and complex reconstruction and analysis programs to extract the physics out of the comparison between simulated and real data. Your events will be merged with thousands of others, files shipped and stored in one of the major universities computing centers, and analyzed by physicists who are all academic people. No commercial use, no individual property, our science is a fully collective process.
For more information on the possibility to look at the events produced, see the PARTICIPANT resources page.
Distributed computing ?
The amount of data recorded by the ATLAS detector will increase again when, next year, the LHC collider will start for a second 3 year long run. A short video presents the Computing infrastructure at CERN : the capacities were extended recently by the creation of a second center, and both should cope with the expected data rate... But simulation has to be done "elsewhere", that is to say "everywhere" and for many physics processes the precision of the measurement is strongly limited by the number of simulated events. This is why we need you, and why your effort will be fully integrated in our research program.
Impact on society ?
The impact of computing at CERN on society : a "Conference Publique" by Bob Jones, responsible for the Openlab project of the CERN IT department, introduces Computing 25 years after the WEB invention. Volunteer computing is mentioned 15 minutes after the beginning, LHC computing on the GRID 15 minutes later, clouds towards the end of the talk... and safety in the discussion with the audience.