arXiv:1707.08514cs.OS
Analyzing IO Amplification in Linux File Systems
Abstract
We present the first systematic analysis of read, write, and space amplification in Linux file systems. While many researchers are tackling write amplification in key-value stores, IO amplification in file systems has been largely unexplored. We analyze data and metadata operations on five widely-used Linux file systems: ext2, ext4, XFS, btrfs, and F2FS. We find that data operations result in significant write amplification (2-32X) and that metadata operations have a large IO cost. For example, a single rename requires 648 KB write IO in btrfs. We also find that small random reads result in read amplification of 2-13X. Based on these observations, we present the CReWS conjecture about the relationship between IO amplification, consistency, and storage space utilization. We hope this paper spurs people to design future file systems with less IO amplification, especially for non-volatile memory technologies.
Ext3, ext4, and XFS have a journal that's constantly being overwritten. On Btrfs, the file system is the journal. It does have a bit of a wandering trees problem [1], where F2FS expressly intends to reduce the wandering tree problem.[2] There's also different approaches that…
cmurfView on HNParent context
Have a look at this paper [0] While there are big gaps of write amplification for metadata writes, on macro benchmarks all filesystems have similar results. btrfs has the biggest WAF, but you can enable compression globally and I suspect that difference alone will make it come…
kasabaliView on HNParent context