Computing Challenges at the LHC

The Large Hadron Collider(LHC) produces about 100 terabytes of raw digitized data every second coming out of the detectors, due to the very high proton-proton collision rate. However, out of billions of collisions, only a tiny fraction is interesting enough to keep for further analysis. This amounts to a total of about 25 petabytes of raw data to be stored and globally distributed per year. Also various forms of processed data, more suitable for analysis, are produced and shared. Thousands of physicists analyze petabytes of data globally every day which is equivalent to the data managed daily by Google or telecommunication giants like AT&T. Globally distributed computing is the only way to handle such an enormous amount of information. The computing needs at the LHC have significantly pushed the envelope of modern day computing. The resulting system, the worldwide LHC Grid Computing (WLCG), is the world's largest computing grid, connecting over 170 computing centers across 36 countries in a truly global network. We shall discuss about how the WLCG evolved through challenges, it's current status, and future direction. The role of emerging technologies like cloud computing in the context of WLCG will also be addressed. We shall also briefly review the impact of Grid computing on Indian academic computing and networking.