Grid and Computing
02 May 2017



Computing Activities in the Particle Physics Department




​Grid Computing

What is a Grid?

A Grid is a way of connecting computers together to allow the sharing of both computer processing power and data storage. A user may submit a job, and software (called middleware) will check the users credentials and find a suitable computer on which to run the job. This computer may be anywhere in the world, provided it is part of the grid infrastructure to which the user has submitted the job. More explanation of what a Grid is is available from the CERN Computing pages.​

The Large Hadron Collider LHC at CERN near Geneva is the largest scientific instrument on the planet. Before it switched on, the LHC was expected to produce roughly 15 Petabytes (15 million Gigabytes) of data annually, but by 2016 the LHC was exceeding both it's design luminosity​ and duty cycle, generating over 30PB of data a year. Data which thousands of scientists around the world need to access and analyse. Grid Computing was chosen as the solution for the analysis of LHC data by the Particle Physics community. The Particle Physics Department was instrumental in this decision from the beginning and has continued to play a major role in the management, development, deployment and maintenance of Grid technology.

Within the UK most of the work is carried out as part of the UK Computing for Particle Physics GridPP project and the European Grid Infrastructure EGI project.

Grid deployment

The Worldwide LHC Computing Grid WLCG project is building and maintaining a data storage and analysis infrastructure for the entire high energy physics community that will use the LHC.

The data from the LHC experiments is distributed around the globe, according to a four-tiered model. A primary backup is recorded on tape at CERN, the "Tier-0" centre of WLCG. After initial processing, this data is distributed to a series of Tier-1 centres, large computer centres with sufficient storage capacity and with round-the-clock support for the Grid. The Tier-1 centres make data available to Tier-2 centres, each consisting of one or several collaborating computing facilities, which can store sufficient data and provide adequate computing power for specific analysis tasks. Individual scientists access these facilities through Tier-3 computing resources, which can consist of local clusters in a University Department or even individual PCs, and which may be allocated to LCG on a regular basis.

GridPP has built a distributed computing Grid across the UK primarily for the use of particle physicists, this forms the UK part of the WLCG. In the United Kingdom there are 4 Tier 2 facilities, NorthGrid, SouthGrid, ScotGrid and London Grid. The Particle Physics Department hosts SouthGrid facilities: which is distributed between the Universities of Bristol, Birmingham, Cambridge, Oxford and Sussex and the Rutherford Appleton Laboratory. The department is also one of 17 Tier 3 facilities within the United Kingdom. Staff in the Particle Physics department also contribute to the deployment and support for the UK Tier 1 centre at RAL. The combined Tier 2/3 in the department currently provides just over 4,000 CPU cores and 3,000 TB of data storage to local and remote researchers, principally supporting the Atlas, CMS and LHCb experiments but also supporting more than 20 smaller communities (Virtual Organisations in grid terminology).

Grid security

In any computing system security is an important concern. In the Grid environment, security poses some additional​ problems. In EGI the Security Co-ordination Group SCG is responsible for the overall security co-ordination.

The EGI​ Grid Security Vulnerability Group SVG, and The EGEE/WLCG Joint Security Policy Group SPG are co-funded by EGEE and GridPP, and are lead by members of the Particle Physics Department.

Management, documentation and user support

Members of the Particle Physics department are part of the management board of the GridPP project, and support the CERN experiments use of the RAL tier 1 as co-ordinators for ATLAS, CMS and LHb.