(More) lessons learnt from the deployment of production worldwide grid services for the Large Hadron Collider at CERN

This paper describes the main lessons learned in deploying production services for the Large Hadron Collider (LHC) at CERN. We cover the principle difficulties faced in bringing this multi-grid service to full production status and how they were resolved. We also list remaining challenges, how they will be addressed, as well as possible future directions for long-term sustainable multi-disciplinary e-infrastructures. Finally, we introduce a simple classification of applications that use - or could use - grid infrastructures and examine how they could coexist and the inherent mutual benefits.