Add to ORKGAFS writes are slow. Part of the reason they appear slow is that opportunities to overlap writing with computation are often not used. This paper describes some of these lost opportunities, and advocates (with some gestures toward objectivity) changes to the AFS cache manager to take advantage of them.http://deepblue.lib.umich.edu/bitstream/2027.42/107963/1/citi-tr-92-7.pd
Improvements in the processing speed of multiprocessors are outpacing improvements in the speed of d...
Journal ArticleFor a parallel architecture to scale effectively, communication latency between proce...
We introduce a simple sequential write benchmark and use it to improve the Linux NFS client's write ...
The AFS Cache Manager fetches files from the AFS file server, and caches them into a local file syst...
Delayed write in a multilevel file system cache hierarchy offers a way to improve performance in div...
We describe two enhancements to the AFS distributed file system intended to improve the performance ...
This paper describes current and anticipated work at the University of Michigan's Center for Informa...
University (CMU) in the 1980’s [H+88]. Led by the well-known Profes-sorM. Satyanarayanan of Carnegie...
Distributed file systems will need new design concepts in order to offer high performance. Automatic...
Fast file systems are critical for high-performance scientific computing, since many scientific appl...
An important design goal of a distributed file system, a component of many distributed systems, is t...
This thesis explores ways in which intermediate cache servers affect the performance and scalability...
This thesis describes the effect of write caching on overall file system performance. It will show t...
An important trend in the design of storage subsystems is a move toward direct network attachment. N...
Current network file system protocols rely heavily on a central server to coordinate file activity a...
Improvements in the processing speed of multiprocessors are outpacing improvements in the speed of d...
Journal ArticleFor a parallel architecture to scale effectively, communication latency between proce...
We introduce a simple sequential write benchmark and use it to improve the Linux NFS client's write ...
The AFS Cache Manager fetches files from the AFS file server, and caches them into a local file syst...
Delayed write in a multilevel file system cache hierarchy offers a way to improve performance in div...
We describe two enhancements to the AFS distributed file system intended to improve the performance ...
This paper describes current and anticipated work at the University of Michigan's Center for Informa...
University (CMU) in the 1980’s [H+88]. Led by the well-known Profes-sorM. Satyanarayanan of Carnegie...
Distributed file systems will need new design concepts in order to offer high performance. Automatic...
Fast file systems are critical for high-performance scientific computing, since many scientific appl...
An important design goal of a distributed file system, a component of many distributed systems, is t...
This thesis explores ways in which intermediate cache servers affect the performance and scalability...
This thesis describes the effect of write caching on overall file system performance. It will show t...
An important trend in the design of storage subsystems is a move toward direct network attachment. N...
Current network file system protocols rely heavily on a central server to coordinate file activity a...
Improvements in the processing speed of multiprocessors are outpacing improvements in the speed of d...
Journal ArticleFor a parallel architecture to scale effectively, communication latency between proce...
We introduce a simple sequential write benchmark and use it to improve the Linux NFS client's write ...