NSF Award
9610104
The goal of this project, funded as a Research Planning Grant (RPG) is to perform preliminary studies and develop a long term research proposal to investigate strategies for implementation of file system extensions in widely used operating systems. Examples of extensions under consideration include disconnected operation, weak and partial connectivity, informed prefetching and file reference tracing. The major difficulty in transferring these technologies into commercial practice stems from frequent inability to modify the native system source code. Therefore, alternative mechanisms and tools, such as microkernel augmented DLLs, installable file systems, loadable modules, proxies and TSR programs, will be considered. The scope of the study will span the families of all widely used operating systems such as Windows family, Unix family, and OS/2 family. In addition to the inability to modify the native system source code, constraints such as hardware limitations, and the effect of the extension on system performance will be investigated. The long term goal of this research is to develop a set of guidelines for practical implementation of file system extensions, and study their impact on the native system, thereby furnishing a vital missing link in the transfer of file system research results into computer practice.***
