The LHCb Grid Simulation: Proof of Concept

M Hushchyn,K Arzymatov,S Roiser,A Baranov,A Ustyuzhanin

doi:10.1088/1742-6596/898/5/052020

M Hushchyn, K Arzymatov + Show 3 more

Open Access

https://doi.org/10.1088/1742-6596/898/5/052020

Copy DOI

Abstract

The Worldwide LHC Computing Grid provides access to data and computational resources to analyze it for researchers with different geographical locations. The grid has a hierarchical topology with multiple sites distributed over the world with varying number of CPUs, amount of disk storage and connection bandwidth. Job scheduling and data distribution strategy are key elements of grid performance. Optimization of algorithms for those tasks requires their testing on real grid which is hard to achieve. Having a grid simulator might simplify this task and therefore lead to more optimal scheduling and data placement algorithms. In this paper we demonstrate a grid simulator for the LHCb distributed computing software.

Highlights

In this paper we demonstrate a grid simulator for the LHCb distributed computing software
The Worldwide LHC Computing Grid (WLCG) [1] is a global collaboration of computing centers which provides computing resources to store, distribute and analyse the data produced by the Large Hadron Collider (LHC) as well as corresponding simulated data
The Tier 0 is the CERN Data Center, which is located in Geneva, Switzerland and at the Wigner Research Center for Physics in Budapest, Hungary

Summary

Introduction

The Worldwide LHC Computing Grid (WLCG) [1] is a global collaboration of computing centers which provides computing resources to store, distribute and analyse the data produced by the Large Hadron Collider (LHC) as well as corresponding simulated data. The Worldwide LHC Computing Grid provides access to data and computational resources to analyze it for researchers with different geographical locations. Job scheduling and data distribution strategy are key elements of grid performance. Having a grid simulator might simplify this task and lead to more optimal scheduling and data placement algorithms.

Results

Conclusion