Hi Matty, Kernel async I/O just works on RAW devices at the moment.
Kernel async I/O give a small incremental improvement, but only in the order of a few percent in optimal cases. The user-land thread emulation works quite well when direct I/O is enabled (without it, the single writer lock will allow only one I/O into a file at a time, circumventing any async advantages). There is likely strong interest to extend kaio to the kernel when there is a file system on the other side which can exploit the list/batch nature of the I/O's. UFS won't likely ever benefit, but other file systems like NFS or ZFS could in the future... Regards, Richard. On Wed, Aug 10, 2005 at 03:46:23PM -0400, Matty wrote: > > Howdy, > > Does anyone know why kernel asynchronous I/O is limited to raw devices? Is > it possible to extend kaio to the file system layer? Would there be any > benefit? Is there a noticeable performance difference between the userland > and kernel asynchronous I/O implemenation when UFS is used with large > Oracle OLTP workloads? Curious to see what folks have seen/measured. > > Thanks for any insight, > - Ryan > _______________________________________________ > perf-discuss mailing list > perf-discuss@opensolaris.org -- :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: Richard Mc Dougall : [EMAIL PROTECTED] Performance and Availability : x31542 Engineering : http://blogs.sun.com/rmc Sun Microsystems Inc : +1 650 352 6438 :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: _______________________________________________ perf-discuss mailing list perf-discuss@opensolaris.org
