Authors: Dheeraj Bhardwaj, Manish K. Sinha
Addresses: Laing O'Rourke Plc, Bridge Place, Crossways, Dartford, Kent, DA2 6SN, UK. ' GridLogics Technologies (P) Ltd., Pune IT Park, 34 Aundh Road, Bhau Patil Marg, Pune 411 020, India
Abstract: I/O has always been performance bottleneck for applications running on clusters. Most traditional storage architectures fail to meet the requirement of concurrent access to the same file that is posed by most High-Performance Computing (HPC) applications. Many parallel and cluster file systems are still plagued by metadata overheads and associated management complexities that prevail in read/write intensive scenarios. In this paper we introduce GridFS, a next generation I/O solution that can scale to hundreds or thousands of nodes and several hundreds of terabytes of storage with very high I/O and metadata throughput. It is based on the Object-based Storage Architecture (OSA) model and goes a step further to eliminate runtime file access overheads as compared to other implementations on the same model.
Keywords: parallel file systems; cluster file systems; object-based storage architecture; OSA; object based storage devices; parallel I/O; input-output; cache coherency; RAID5; grid computing; clusters; high performance computing; runtime file access overheads.
International Journal of Computational Science and Engineering, 2006 Vol.2 No.5/6, pp.287 - 291
Published online: 08 Aug 2007 *Full-text access for editors Access for subscribers Purchase this article Comment on this article