Data Structures for Storing File Namespace in Distributed File System

The distributed file system (DFS) provides a mechanism where a file can be stored across several physical computer nodes ensuring replication transparency and failure transparency. In order to achieve this, one important feature of DFS's data structure is to employ a proper way to organize the large scale data and file namespaces to provide a fast access speed. In order to analyze file namespace issue, we present the comparison among three major data structures, AVL, Red-Black (RB) and B trees which are popular for organizing file namespace for DFS. This paper also presents significant results obtained from our intensive experiments. From the results collected, we conduct a precise vision about the performance comparison of these data structures and propose that AVL tree is the most appropriate solution.