As storage systems grow larger and more complex, the traditional block-based file systems cannot satisfy the large workload. More recent distributed file systems have adopted architectures based on object-based storage. This paper presents a framework of efficient storage management for distributed storage system. In object storage side, low-level storage tasks and data distribution must be managed and in metadata server side, we will manage how to scale the metadata. Due to the high space efficiency and fast query response, bloom filters have been widely utilized in recent storage systems. So, we will also utilize bloom-filter based approach to manage metadata by taking the advantages of bloom-filter and the semantic-based scheme will also be used to narrow the managed workload. In this paper, we will neglect the data distribution of object-based storage side.

Date Published: 13 January 2012
PDF: 5 pages
Proc. SPIE 8350, Fourth International Conference on Machine Vision (ICMV 2011): Computer Vision and Image Analysis; Pattern Recognition and Basic Technologies, 83502C (13 January 2012); doi: 10.1117/12.920129

Published in SPIE Proceedings Vol. 8350:
Fourth International Conference on Machine Vision (ICMV 2011): Computer Vision and Image Analysis; Pattern Recognition and Basic Technologies
Safaa S. Mahmoud; Zhu Zeng; Yuting Li, Editor(s)