{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

File system design is 99 block allocation mcvoy

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ile layout. “File system design is 99% block allocation.” [McVoy] Competing goals for block allocation: • allocation cost • bandwidth for high-volume transfers • efficient directory operations Goal: reduce disk arm movement and seek overhead. Misc Misc Improvements: Defrag, Journalling Defrag, Defragmentation can be done periodically to improve performance Journalling Journalling – write a log entry to disk before any operations are performed – after a crash re-execute log e.g. file delete – 3 ops ops must be idempotent (executed repeatedly) Improve Improve Bandwidth How? Reduce Reduce Seek Penalty How can we do this? Ideas? Log Log-structured File Systems (up to recovery) Motivation Motivation Processor speeds have been following Moore’s Law Same for main memory size and disk capacity Disk access times have evolved much more slowly Consequences Consequences Larger memory sizes mean larger caches • caches can capture most read accesses • disk traffic will be dominated by writes • caches can also act as write buffers replacing many small writes by fewer bigger writes Key issue is to increase disk write performance by eliminating seeks (for small files) Workload Workload considerations Disk system performance is strongly affected by workload Many workloads are dominated by accesses to small files • enterprise: document editing, e.g. • many random disk accesses (across different files) • file creation and deletion times dominated by directory and inode updates • sever...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online