From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Mason Subject: Re: interesting use case for multiple devices and delayed raid? Date: Thu, 02 Apr 2009 07:27:16 -0400 Message-ID: <1238671636.27540.8.camel@think.oraclecorp.com> References: <1238577447.9099.150.camel@pc.interlinx.bc.ca> <3a7f57190904010313q230f004chf7db770bb1169daa@mail.gmail.com> <3a7f57190904012241y40f7a906x3edd1ad8002c3208@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain Cc: linux-btrfs@vger.kernel.org To: Dmitri Nikulin Return-path: In-Reply-To: <3a7f57190904012241y40f7a906x3edd1ad8002c3208@mail.gmail.com> List-ID: On Thu, 2009-04-02 at 16:41 +1100, Dmitri Nikulin wrote: > On Thu, Apr 2, 2009 at 8:04 AM, Brian J. Murrell wrote: > >> A more complete solution, that requires no software changes, would be > >> to have 3 or 4 disks. A stripe for really fast reads and writes, and > >> another disk (or another stripe) to act as a slave to the data being > >> written to the primary stripe. This seems to do what you want, at a > >> small price premium. > > > > No. That's not really what I am describing at all. > > Well you get the bandwidth of 2 disks when reading and writing, and > still mirrored to a second stripe as time permits. Kind of like > delayed RAID10. > > > I apologize if my original description was unclear. Hopefully it is > > more so now. > > Yes. It'll be up to the actual filesystem devs to weigh in on whether > it's worth implementing. > It's an interesting idea, but I think we've got fast front end devices higher up on the todo list. That will still support the destaging to slower disks idea, but will be more flexible overall. -chris