지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학위논문
- 저자정보
- 지도교수
- 서의성
- 발행연도
- 2016
- 저작권
- 성균관대학교 논문은 저작권에 의해 보호받습니다.
이용수0
이 논문의 연구 히스토리 (2)
- 이 논문의 후속연구가 궁금하신가요?
- 연관 학술논문 또는 학술발표를 통해 보다 발전된 연구결과를 확인하실 수 있습니다.
- 이 논문의 연구 히스토리 확인하기
초록· 키워드
오류제보하기
File fragmentation can be caused by various reasons such as frequent small file creation and deletion. This file fragmentation is commoner on Log structured File System than other file systems. When many processes issue write operations alternately on a log-structured file system (LFS), the created files can be fragmented on the file system layer although LFS sequentially allocates new blocks to each process. Unfortunately, file fragmentation degrades read performance because the fragmented files cause block I/Os to be split and decrease sequentiality of block I/Os. But prior schemes about file defragmentation make additional block I/Os to rearrange fragmented blocks. It is harmful to flash memory, which has limited lifetime. In this paper, we suggest a sorting cleaning on LFS, which eliminates file fragmentation. During a cleaning process of LFS, our sorting cleaning process rearranges valid data blocks, with sorting by inode numbers before the cleaning process copies the valid blocks to a new segment. This sorting process relocates fragmented blocks contiguously. To exploit our method, we also suggest fragmentation-aware victim selection policy. It selects a highly fragmented segment as a victim segment utilizing fragmentation degree of segments. Because fragmented files can be de-fragmented by sorting cleaning, this policy improves the sorting cleaning effect. In our experiments, our cleaning scheme eliminates 98.5% of file fragmentation as compared to file fragmentation before cleaning at most. Consequently, our cleaning method improves sequential read throughput by 677% at most.
목차
Abstract 11. Introduction 32. Related works and motivation 73. Log-structured file system 124. LFS cleaning scheme for file defragmentation 184.1 Sorting cleaning for file defragmentation 194.2 Fragmentation-aware victim selection policy 215. Evaluation 255.1 Experimental setup 255.2 Results 265.2.1 Sorting cleaning 265.2.2 Fragmentation-aware victim selection policy 316. Conclusion 33References 34Korean abstract 36
최근 본 자료
댓글(0)
0