2011 | OriginalPaper | Chapter
Efficient De-Fragmented Writeback for Journaling File System
Authors : Seung-Ho Lim, Hyun Jin Choi, Jong Hyuk Park
Published in: Secure and Trust Computing, Data Management and Applications
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Journaling file systems are widely used file systems to guarantee data integrity. However, the system performance is degraded due to data request fragmentations. We propose a technique that efficiently handles fragmented data workloads in journaling file system, which we call De-Fragmented Writeback(DFW). The first method of the DFW sorts write orders of the atomic data set in accordance with their disk block numbers before issuing into write requests. The second method of the DFW searches for data fragments in the sorted data set and tries to fill up holes between adjacent data fragments using dirty blocks in main memory. The filling of holes between data fragments converts fragmented data blocks into sequential ones so that this method lowers the number of write requests and reduces unnecessary disk-head movements to write the blocks.