Distributed development of large-scale global systems: exploring the collaborative practices of particle physicists as they develop a grid for the LHC

Grid computing promises to distribute and share computing resources “on tap” and provide transparent communication and collaboration between virtual groups (Foster and Kesselman 2003). Yet developing and implementing such complex information infrastructures requires collaboration among a range of dispersed groups, and flexibility and adaptability to volatile requirements (Berman, Geoffrey et al. 2003). Here, we examine a case-study of Grid development within particle physics, the LCG (Large hadron collider (LHC) Computing Grid), in an attempt to explore how such a large-scale distributed system is developed collaboratively in a global way in readiness for the launch of the LHC at CERN in September 2008. The particle physics community is well-known for the development of other cutting edge distributed systems (notably the web) and is itself highly distributed, so presenting a context where distinctive collaborative practices emerge.