Communications of the ACM
The use of name spaces in Plan 9
ACM SIGOPS Operating Systems Review
Accessing Files in an Internet: The Jade File System
IEEE Transactions on Software Engineering
The MOSIX multicomputer operating system for high performance cluster computing
Future Generation Computer Systems - Special issue on HPCN '97
Communications of the ACM - Special 25th Anniversary Issue
BProc: the Beowulf distributed process space
ICS '02 Proceedings of the 16th international conference on Supercomputing
The LOCUS distributed operating system
SOSP '83 Proceedings of the ninth ACM symposium on Operating systems principles
MRNet: A Software-Based Multicast/Reduction Network for Scalable Tools
Proceedings of the 2003 ACM/IEEE conference on Supercomputing
Versatility and Unix semantics in namespace unification
ACM Transactions on Storage (TOS)
TCON'95 Proceedings of the USENIX 1995 Technical Conference Proceedings
Lessons learned at 208K: towards debugging millions of cores
Proceedings of the 2008 ACM/IEEE conference on Supercomputing
Hi-index | 0.00 |
Group file operations enable tools and middleware to operate upon a large group of files located across thousands of independent servers in a scalable fashion, a necessary requirement for effective use of today's largest distributed systems. Our initial prototype of group file operations showed scalability benefits for several tools, but also revealed the importance of having a scalable method for defining useful groups. We have developed a language called FINAL for describing name space composition in a flexible and scalable manner. Clients of our TBON-FS distributed file system can use this language to compose a single-system image (SSI) name space that automatically creates useful groups in a scalable fashion. We provide many examples of traditional and SSI name space compositions that can be described using FINAL, and report how TBON-FS can compose a global name space from tens of thousands of independent name spaces in one-quarter second.