File systems are key operating system components because they store and retrieve files as needed. Traditional hierarchical file systems no longer suit current users’ expectations in organizing extensive collections of files for easy retrieval based on intrinsic and user-defined properties. The number of files in users’ collections is growing substantially, partly because of the ease with which consumer devices capture information. With the enormous capacity of contemporary storage devices and the rising trend of users storing their data in the cloud, they only fuel the number of files needing to be managed. We suggest that current file systems require an enhanced technique of data organization and access so that users can properly handle these ever-increasing data collections. File systems are key operating system components because they store and retrieve files as needed. This study offers improvements to the conventional Hierarchical File System to improve file organization and retrieval through built-in querying capabilities and support filesystem-level operations that execute significant amounts of metadata updates. This is accomplished using attributes (name-value pairs) in a file collection hierarchy. A series of improvements to the HFS introduce the “AttFS” file system. These improvements include using attributes rather than names, logical collections rather than directories, and introducing a query language to the API. We assess the expressive capacity of the resulting model, demonstrate that it solves the relevant shortcomings of traditional file systems in this space, and compare our new approach to those provided by others and our earlier work. We conclude that attributes are better suited than tags to overcome traditional HFS shortcomings.
