btrFS

We have just started testing Open Suse 13.2 and perhaps this is the time to start evaluating performance on this filesystem. Has anyone done any testing with this? Any whitepapers out there? I read somewhere that ext3 is quicker than ext4 and ext2 is quicker than ext3. Our current choice is ext3 for chunk partitions using DIRECT_IO=1. We do have some customers using raw but less so now due to ease of administration.

Stuart: EXT3 with journaling disabled is the same, performance wise, as EXT2 with the advantage of being expandable without having to unmount it. Journaling is not good for databases. EXT4 is unsafe. Let us know how your testing goes. Here is a link to a performance test posted online comparing EXT3, EXT4, XFS, IFS, and btrFS in several option configurations: http://btrfs.boxacle.net/ Here are the results for one of the tests that is relevant to database performance - 4K random writes on a 100MB file: http://btrfs.boxacle.net/repository/raid/Initial-compare/Initial-Compare-RAID0_L arge_file_random_writes._num_threads=16.html Bottom line? btrFS performance on random writes is awful. Best performance for btrFS was ~30MB/s versus ~130MB/s for EXT3. In IOPS btrFS managed about 7500 ops/s versus EXT3 at ~33,000 ops/s Here's another that shows btrFS in better light, but in the mySQL test, the closest to Informix in behavior, btrFS is slowest: http://www.ilsistemista.net/index.php/linux-a-unix/21-ext3-ext4-xfs-and-btrfs-fi lesystems-comparison-on-linux-kernel-300.html?limitstart=0 Art Art S. Kagel, President and Principal Consultant ASK Database Management www.askdbmgt.com Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Wed, Nov 19, 2014 at 5:44 AM, STUART STEPHENS < stuart.stephens@monitorsoft.com> wrote: > We have just started testing Open Suse 13.2 and perhaps this is the time to > start evaluating performance on this filesystem. > > Has anyone done any testing with this? Any whitepapers out there? > > I read somewhere that ext3 is quicker than ext4 and ext2 is quicker than > ext3. > Our current choice is ext3 for chunk partitions using DIRECT_IO=1. We do > have > some customers using raw but less so now due to ease of administration. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --001a11c33e908ce30e0508355615

Nice info, Art - thanks! -- Regards Spokey > On 19 Nov 2014, at 12:27, Art Kagel <art.kagel@gmail.com> wrote: > > Stuart: > > EXT3 with journaling disabled is the same, performance wise, as EXT2 with > the advantage of being expandable without having to unmount it. Journaling > is not good for databases. EXT4 is unsafe. > > Let us know how your testing goes. > > Here is a link to a performance test posted online comparing EXT3, EXT4, > XFS, IFS, and btrFS in several option configurations: > > http://btrfs.boxacle.net/ > > Here are the results for one of the tests that is relevant to database > performance - 4K random writes on a 100MB file: > > > http://btrfs.boxacle.net/repository/raid/Initial-compare/Initial-Compare-RAID0_L arge_file_random_writes._num_threads=16.html > > Bottom line? btrFS performance on random writes is awful. Best > performance for btrFS was ~30MB/s versus ~130MB/s for EXT3. In IOPS btrFS > managed about 7500 ops/s versus EXT3 at ~33,000 ops/s > > Here's another that shows btrFS in better light, but in the mySQL test, the > closest to Informix in behavior, btrFS is slowest: > > > http://www.ilsistemista.net/index.php/linux-a-unix/21-ext3-ext4-xfs-and-btrfs-fi lesystems-comparison-on-linux-kernel-300.html?limitstart=0 > > Art > > Art S. Kagel, President and Principal Consultant > ASK Database Management > www.askdbmgt.com > > Blog: http://informix-myview.blogspot.com/ > > Disclaimer: Please keep in mind that my own opinions are my own opinions > and do not reflect on the IIUG, nor any other organization with which I am > associated either explicitly, implicitly, or by inference. Neither do > those opinions reflect those of other individuals affiliated with any > entity with which I am affiliated nor those of the entities themselves. > > On Wed, Nov 19, 2014 at 5:44 AM, STUART STEPHENS < > stuart.stephens@monitorsoft.com> wrote: > >> We have just started testing Open Suse 13.2 and perhaps this is the time to >> start evaluating performance on this filesystem. >> >> Has anyone done any testing with this? Any whitepapers out there? >> >> I read somewhere that ext3 is quicker than ext4 and ext2 is quicker than >> ext3. >> Our current choice is ext3 for chunk partitions using DIRECT_IO=1. We do >> have >> some customers using raw but less so now due to ease of administration. >> >> >> >> > ******************************************************************************* >> Forum Note: Use "Reply" to post a response in the discussion forum. >> >> > > --001a11c33e908ce30e0508355615 > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. >