On Fri, Oct 10, 2008 at 11:03 PM, Bob Friesenhahn <[EMAIL PROTECTED]> wrote: > On Fri, 10 Oct 2008, Marion Hakanson wrote: >> >> Yes, this advice was unexpected to me. Slicing up a spinning disk drive >> is going to cause lots of extra seeks, but I'm having a hard time imagining >> that an SSD is going to even notice that it's been sliced up. > > >From what I have read, for most SSDs, if an already written part of an > SSD needs to be re-written, the previously existing part needs to be > read into the SSD's cache, 512K of the SSD is erased, the SSD cache is
"Block" (erasable unit) size depends on the manufacturer, and AFAIK 256K has been quoted most often. Other than that the comment is exactly right: FTL's wear leveling algorithm will effectively interleave i/o to both partitions, and I doubt this is good behavior for a ZIL device. Regards, Andrey > updated to reflect the updated bits, and then the 512K is restored. > This could be a consideration if there may be multiple overlapping > write requests. The SSD itself does not know about partitioning and > in fact tries to distribute the write load as evenly as possible as > part of the wear leveling algorithm. > > Bob > ====================================== > Bob Friesenhahn > [EMAIL PROTECTED], http://www.simplesystems.org/users/bfriesen/ > GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ > > _______________________________________________ > storage-discuss mailing list > [email protected] > http://mail.opensolaris.org/mailman/listinfo/storage-discuss > _______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
