Speaker
Dr.
Maria Girone
(CERN)
Description
The CMS Computing system was successfully commissioned and operated in the first run of LHC. Beginning in 2015, CMS will collect, process, simulate and analyze 1kHz of higher complexity, higher energy events. In order to meet this increased computing challenge within the resource budget expected, we have had to evolve the computing model and the techniques used. In this presentation we will discuss the challenges expected in Run2 and the choices made to try to mitigate them. We will outline the improvements in the processing system and the goal to maximize flexibility in how processing resources are used. We will present how the primary reconstruction will be distributed between CERN and the global network of Tier-1 processing centers, and we will show the progress made in allowing resources to be shared across workflows. We will show the development and deployment status of the CMS data federation and how this is expected to reduce the growth of disk resources needed to facilitate analysis, and we will present the progress on a new analysis submission tool for CMS.
Primary authors
Ian Fisk
(FNAL)
Dr.
Maria Girone
(CERN)
