One thing that comes to mind is partition alignment. Have you made sure that the partition starts at a filesystem block size boundary? Check with fdisk -lu
Thanks, Herbert. On 4/12/13 8:43 AM, Andy wrote: > I previously wrote about slowdowns during writing and deleting files at > the same time on an OCFS2 volume. I've tried writeback, and it still > slows down a lot. > > I have also been doing some testings on a 2TB ocfs2 volume on a SSD. I > was wondering how much better it would perform on an SSD. Unfortunately, > I have run into an issue on the SSD, where ocfs2 ends up doing a whole > ton of reads from the volume, and never gets much of a chance to write. > This happens when I access the volume from 2 systems at the same time. > My test scripts just use dd to write various sized files to various > directories. Why would ocfs2 start doing a whole bunch of reads? > > Thanks, > > Andy > > _______________________________________________ > Ocfs2-users mailing list > Ocfs2-users@oss.oracle.com > https://oss.oracle.com/mailman/listinfo/ocfs2-users _______________________________________________ Ocfs2-users mailing list Ocfs2-users@oss.oracle.com https://oss.oracle.com/mailman/listinfo/ocfs2-users
