This paper reports on the effectiveness of the caching strategy used in the Sprite network file system based on data taken over several weeks of day-to-day usage by a variety of users. Measurements include cache consistency activity, long term I/O traffic rates, long term cache hit rates, and the averages and variations in the size of the variable-sized caches. Network traffic is compared with traffic to the local cache, and the effects of paging traffic are con-sidered. The overall conclusion is that the caching system is quite effective and poses a low overhead. Using a delayed write strategy, 40 % to 50 % of the data written to client caches is never written through to a server, and less than 1 % of the open operations by clients resulte...
We assembled 18 months of transfer logs from a production High Performance Storage System (HPSS) sys...
Routing algorithms are known to be potential bottlenecks for packet processing. Network ow caching c...
Fast Internet content delivery relies on two layers of caches on the request path. Firstly, content ...
The Sprite network operating system uses large main-memory disk block caches to achieve high perform...
The Sprite network operating system uses large main-memory disk block caches to achieve high perform...
This thesis describes the effect of write caching on overall file system performance. It will show t...
Caching has long been recognized as a powerful performance enhancement technique in many areas of co...
This thesis explores ways in which intermediate cache servers affect the performance and scalability...
This paper examines the behavior of mechanisms for providing variable-size file data caches. It pres...
Current network file system protocols rely heavily on a central server to coordinate file activity a...
Large scientific collaborations often have multiple scientists accessing the same set of files while...
This paper presents the results of simulating file name and attribute caching on client machines in ...
This paper presents the results of simulating file name and attribute caching on client machines in ...
The Internet Domain Name System (DNS) is a distributed, hierarchical and flexible system for the map...
The caching of data and/or locks at client workstations is an effective technique for improving the ...
We assembled 18 months of transfer logs from a production High Performance Storage System (HPSS) sys...
Routing algorithms are known to be potential bottlenecks for packet processing. Network ow caching c...
Fast Internet content delivery relies on two layers of caches on the request path. Firstly, content ...
The Sprite network operating system uses large main-memory disk block caches to achieve high perform...
The Sprite network operating system uses large main-memory disk block caches to achieve high perform...
This thesis describes the effect of write caching on overall file system performance. It will show t...
Caching has long been recognized as a powerful performance enhancement technique in many areas of co...
This thesis explores ways in which intermediate cache servers affect the performance and scalability...
This paper examines the behavior of mechanisms for providing variable-size file data caches. It pres...
Current network file system protocols rely heavily on a central server to coordinate file activity a...
Large scientific collaborations often have multiple scientists accessing the same set of files while...
This paper presents the results of simulating file name and attribute caching on client machines in ...
This paper presents the results of simulating file name and attribute caching on client machines in ...
The Internet Domain Name System (DNS) is a distributed, hierarchical and flexible system for the map...
The caching of data and/or locks at client workstations is an effective technique for improving the ...
We assembled 18 months of transfer logs from a production High Performance Storage System (HPSS) sys...
Routing algorithms are known to be potential bottlenecks for packet processing. Network ow caching c...
Fast Internet content delivery relies on two layers of caches on the request path. Firstly, content ...