Granularity and semantic level of replication in the Echodistributed file system
Hisgen, A.
Birrell, A.
Jerian, C.
Mann, T.
Schroeder, M.
Swart, G.
DEC Syst. Res. Center, Palo Alto, CA;
This paper appears in: Management of Replicated Data, 1990. Proceedings., Workshop on the
Publication Date: 8-9 Nov 1990
On page(s): 2-4
Meeting Date: 11/08/1990 - 11/09/1990
Location: Houston, TX, USA
ISBN: 0-8186-2085-4
References Cited: 9
INSPEC Accession Number: 3944602
Digital Object Identifier: 10.1109/MRD.1990.138234
Current Version Published: 2002-08-06
Abstract
The choices for semantic level and granularity that have been made
in the Echo distributed file system are examined. The primary goals of
Echo are to explore issues of scaling, availability, and performance.
For scaling and uniformity of access, Echo provides a global,
hierarchical name space. Replication is used for availability.
Performance is achieved by distributed caching on clients and by using a
log on the file server to reduce disk seeks. The log also records
information about updates that are in progress, and this information is
used during crash recovery to bring all replicas into agreement. The
Echo hierarchical name space is structured as a collection of subtrees,
called echo volumes, that are glued together to form a single name
space. Each echo volume may be implemented by a different service, for
example, a name service or a file service. The file service component of
Echo is discussed
Index
Terms
Available to subscribers and IEEE members.
References
Available to subscribers and IEEE members.
Citing Documents
Available to subscribers and IEEE members.
Cart
