On Thu, Feb 8, 2018 at 11:40 AM, Vitaliy Garnashevich < vgarnashev...@gmail.com> wrote:
> Anyway, there are still some strange things happening when > effective_io_concurrency is non-zero. > > ... > > Vitaliy > > I was researching whether I could optimize a concatenated lvm2 volume when I have disks of different speeds (concatenated - not striped - and I think I can if I concatenate them in the right order - still testing on that front), when I came across this article from a few years ago: http://www.techforce.com.br/content/lvm-raid-xfs-and-ext3-file-systems-tuning-small-files-massive-heavy-load-concurrent-parallel In the article he talks about the performance of parallel io on different file systems. Since I am already running XFS that led me to this tunable: http://xfs.org/docs/xfsdocs-xml-dev/XFS_Filesystem_Structure/tmp/en-US/html/Allocation_Groups.html Which brought me back to this discussion about effective_io_concurrency from a couple of weeks ago. I noticed that the recent round of tests being discussed never mentioned the file system used. Was it XFS? Does changing the agcount change the behaviour?