High Energy Physics
Jump to navigation
Jump to search
The High Energy Physics (HEP) HUC represents the 4 LHC experiments at CERN, which
are fully relying on the use of grid computing for their offline data distribution,
processing and analysis. The HEP computing systems are probably the most complex
grid-integrated applications currently in production. The services ran by the HEP HUC
can be classified into:
- Experiment computing systems and services: Software stacks developed by the
experiments on top of the WLCG middleware to implement their particular computing models. Nevertheless, successful examples of experiment computing systems reuse exist, e.g. the Linear Collider Detector (see Section 5.3.2)
- Middleware services: VO independent, high-level grid services
- Data Management: Services for data discovery and data transfer, e.g. LCG File Catalogue, gLite File Transfer Service, WLCG Disk Pool Manager etc. See Section 6.7 for a complete description of each service.
- Workload Management: Services that allow users to submit and manage generic batch jobs on grid resources, e.g. Ganga (see Section 6.2), gLite Workload Management System etc.
- Persistency: Framework to interface database access for storing and retrieving different types of scientific data, such as event and conditions data. [R 4]
- Monitoring: Application and site monitoring to follow the experiment activities and the state of the grid infrastructure respectively, e.g. Dashboards (see Section 6.1), SAM/Nagios monitoring and HammerCloud (see Section 6.6)