Date: Wed, 9 Oct 1996 03:03:34 -0700 (PDT) From: asami@freebsd.org (Satoshi Asami) To: bsdscsi@shadows.aeon.net Cc: pjchilds@imforei.apana.org.au, freebsd-scsi@freebsd.org Subject: Re: striping/mirroring? Message-ID: <199610091003.DAA04113@silvia.HIP.Berkeley.EDU> In-Reply-To: <199610080853.KAA03301@shadows.aeon.net> (message from mika ruohotie on Tue, 8 Oct 1996 10:53:57 %2B0200 (EET))
next in thread | previous in thread | raw e-mail | index | archive | help
* does this work already? (i cant experiment yet, disks are on purchase * list i have to get signed before i get the disks) You mean mirroring? Yes it's working, has been for a few months. * and am i right assuming (some old post listed these) that 128 is the * interleave value that gives me most speed for reads? Depends on what kind of reads you are talking about. For large sequential reads (here the size of individual read()s don't matter -- the "large" refers to the total size that's read in succession) and many disks, something a little smaller is usually better, e.g., 32 or 64. For random reads, it should probably be the size of the read, i.e., if your reads are 16K, then 32 would do the best. (Unless the read sizes are very large, say 1M or so...in which case, treat this as "sequential" as described above.) Of course, I'm assuming the reads occur at offsets that are integer multiples of the read sizes -- if not, I guess you want something a little larger to reduce the chance of a single read falling between two disks. However, since 128 does reasonably well for both reads and writes, that is the size I would recommend for normal (read/write) workloads. * question, do i have to mirror as many drives as i am striping? since i'd * rather strip to 3 and mirror to one... (living close to the edge... =) ) That's not how mirroring works. (Although it is possible to have a 3-disk stripe and one disk for mirroring using a different implementation than our ccd, that won't protect all 3 disks.) You are probably talking about write-ahead logging, which we don't have. When LFS comes back to earth, you may be able to combine ccd and LFS to do something of that sort. * > * Advantages are reasonable speed, and a measure of safety. If one disk * > * fails then you can just "turn-off" ccd and continue on with the other * > * good pair of disks. * * it's this simple? nothing lost? sounds cool... =) That's why you are paying a very high cost on redundancy (50% of diskspace just for mirroring). Of course, it's still cheaper than hardware RAID boxes. :> * but if the spare is online, and running there's the chance it is broken * already... and since it just hangs there i would not know it until... You can use a SCSI command to tell it to stop spinning while it's sitting there as a spare. In which case, I would put the chance of it breaking roughly equal to a disk just waiting in the shipping box. :) Satoshi
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199610091003.DAA04113>