Abstract
Parallel I/O needs keep pace with the demand of high performance computing applications. I/O bottleneck still is one of the key problems which restrict the development of high performance computer. Object-based storage combines the advantages of SAN and NAS, adopts an object-based interface, and has been a new orientation of massive storage. The trend as such makes it very promising to build a scalable I/O system for highperformance computing through implementation of object-based storage system based on high-performance interconnects. Lustre is a high performance distributed object parallel file system, is being widely studied and applied. Exploiting performance of Lustre depends on supports of high performance interconnects. In this paper, firstly we designed and implemented a customized high performance interconnect chip, HSNI (High performance Storage Network Interface), and introduced the characteristics of HSNI. Then we analyzed the communication features which support high performance of Lustre. Experimental results show that HSNI can greatly exert the performance of Lustre and has a very good scalability, so its very suitable for building the high performance object-based storage systems for high performance computing.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Lyman, P., Varian, H.R., et al.: How much information (2003)
Gray, J.: What Next? A Few Remaining Problems in Information Technology (1988)
Wber, R.O.: SCSI Object-Based Storage Device Commands(OSD), Document Number: ANSI/INCITS 400-2004. International Committee for Information Standards (December 2004)
Lustre, http://www.lustre.org
David Nagle, A.M., Serenyi, D.: The Panasas ActiveScale Storage Cluster-Dellivering Scalable High Bandwidth Storage. In: Proceedings of Supercomputing 2004 (2004)
Braam, P.J.: The Lustre storage architecture. Technical report, Cluster File Systems, Inc. (2002)
Rodeh, O., Teperman, A.: zFS- a scalable distributed file system using object disks. In: Proceedings of the 20th IEEE/11th NASA Goddard Conference on Mass Storage Systems and Technologies, pp. 207–218 (2003)
Coda Project, http://www.codaproject.org/
Myrinet, http://www.myri.com/
InfiniBand Architecture Specification, Release1.1. InfiniBand Trade Association (2002)
Petrini, F., Feng, W.C., Hoisie, A., Coll., S., Frachtenberg, E.: The Quadrics network: High-Performance Clustering Technology. IEEE Micro. 22(1) (2002)
Yu, W., Noronha, R., Liang, S., Panda, D.K.: Benefits of High Speed Interconnects to Cluster File Systems: A Case Study with Lustre. In: IPDPS 2006 (2006)
Yu, W., Panda, D.K.: Benefits of Quadrics Scatter/Gather to PVFS2 Noncontiguous IO. In: PACT 2005 (2005)
Thakur, R., Gropp, W., Lusk, E.: On Implementing MPI-IO Portably and with High Performance. In: Proceedings of the 6th Workshop on I/O in Parallel and Distributed Systems, pp. 23–32 (1999)
IOR Benchmark, http://www.llnl.gov/asic/purple/benchmarks/limited/ior
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Guo, Y., Yang, X., Luo, L., Li, Q., Liu, L. (2009). High Performance Support of Lustre over Customized HSNI for HPC. In: Dou, Y., Gruber, R., Joller, J.M. (eds) Advanced Parallel Processing Technologies. APPT 2009. Lecture Notes in Computer Science, vol 5737. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-03644-6_17
Download citation
DOI: https://doi.org/10.1007/978-3-642-03644-6_17
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-03643-9
Online ISBN: 978-3-642-03644-6
eBook Packages: Computer ScienceComputer Science (R0)