Abstract
NFS is a distributed file system which is widely used in UNIX environments. Many studies have been worked to improve its I/O performance and scalability. However, existing architectures, which have either single load balancer or meta server, could suffer from single-point-of-failure and unnecessary network delay. We propose a new NFS architecture called mNFS cluster to improve its scalability. Experimental results show that mNFS outperforms both Linux NFS and NFSp by 64% and 99%, respectively. © Springer-Verlag 2004.
Cite
CITATION STYLE
Lee, W. G., Park, C. I., & Kim, D. W. (2004). mNFS: Multicast-based NFS cluster. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3038, 363–370. https://doi.org/10.1007/978-3-540-24688-6_49
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
