Fido: A Cache That Learns to Fetch
VLDB '91 Proceedings of the 17th International Conference on Very Large Data Bases
Design and Implementation of a Predictive File Prefetching Algorithm
Proceedings of the General Track: 2002 USENIX Annual Technical Conference
The Case for Efficient File Access Pattern Modeling
HOTOS '99 Proceedings of the The Seventh Workshop on Hot Topics in Operating Systems
Using Multiple Predictors to Improve the Accuracy of File Access Predictions
MSS '03 Proceedings of the 20 th IEEE/11 th NASA Goddard Conference on Mass Storage Systems and Technologies (MSS'03)
Group-Based Management of Distributed File Caches
ICDCS '02 Proceedings of the 22 nd International Conference on Distributed Computing Systems (ICDCS'02)
Long Term Distributed File Reference Tracing: Implementation and Experience
Long Term Distributed File Reference Tracing: Implementation and Experience
Performing File Prediction with a Program-Based Successor Model
MASCOTS '01 Proceedings of the Ninth International Symposium in Modeling, Analysis and Simulation of Computer and Telecommunication Systems
Characteristics of File System Workloads
Characteristics of File System Workloads
File access prediction with adjustable accuracy
PCC '02 Proceedings of the Performance, Computing, and Communications Conference, 2002. on 21st IEEE International
An analytical approach to file prefetching
ATEC '97 Proceedings of the annual conference on USENIX Annual Technical Conference
Why does file system prefetching work?
ATEC '99 Proceedings of the annual conference on USENIX Annual Technical Conference
On the design of a new Linux readahead framework
ACM SIGOPS Operating Systems Review - Research and developments in the Linux kernel
File access prediction using neural networks
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
Most existing studies of file access prediction are experimental in nature and rely on trace driven simulation to predict the performance of the schemes being investigated. We present a first order Markov analysis of file access prediction, discuss its limitations and show how it can be used to estimate the performance of file access predictors, such as First Successor, Last Successor, Stable Successor and Best-k-out-of-n. We compare these analytical results with experimental measurements performed on several file traces and find out that specific workloads, and indeed individual files, can exhibit very different levels of non-stationarity. Overall, at least 60 percent of access requests appear to remain stable over at least a month.