Dynamic data replication in LCG 2008

  • Authors:
  • C. Nicholson;D. G. Cameron;A. T. Doyle;A. P. Millar;K. Stockinger

  • Affiliations:
  • University of Glasgow, Glasgow G12 8QQ, U.K.;University of Oslo, P.O. Box 1048, Blindern, N-0316 Oslo, Norway;University of Glasgow, Glasgow G12 8QQ, U.K.;University of Glasgow, Glasgow G12 8QQ, U.K.;Lawrence Berkeley National Laboratory, Berkeley, CA 94720, U.S.A.

  • Venue:
  • Concurrency and Computation: Practice & Experience - UK e-Science All Hands Meeting 2006
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

To provide performance access to data from high-energy physics experiments such as the Large Hadron Collider (LHC), controlled replication of files among grid sites is required. Dynamic, automated replication in response to jobs may also be useful and has been investigated using the grid simulator OptorSim. In this paper, results are presented from simulations of the LHC Computing Grid in 2008, in a physics analysis scenario. These show, first, that dynamic replication does give improved job throughput; second, that for this complex grid system, simple replication strategies such as Least Recently Used and Least Frequently Used are as effective as more advanced economic models; third, that grid site policies that allow maximum resource sharing are more effective; and lastly, that dynamic replication is particularly effective when data access patterns include some files being accessed more often than others, such as with a Zipf-like distribution. Copyright © 2008 John Wiley & Sons, Ltd. (Cameron) Work done while at CERN, the European Organization for Nuclear Research, 1211 Geneva, Switzerland. (Stockinger) Work done while at Lawrence Berkeley National Laboratory, Berkeley, CA 94720, U.S.A.