Date: Wed, 28 Oct 1998 01:38:26 -0600 (CST) From: Joe Greco <jgreco@solaria.sol.net> To: hackers@FreeBSD.ORG Subject: Re: Multi-terabyte disk farm Message-ID: <199810280738.BAA10774@aurora.sol.net> In-Reply-To: <3.0.3.32.19981028011701.0070082c@bugs.us.dell.com> from Tony Overfield at "Oct 28, 98 01:17:01 am"
next in thread | previous in thread | raw e-mail | index | archive | help
> At 11:00 PM 10/27/98 -0600, you wrote: > >Writing to a RAID involves a performance penalty where all drives have to > >participate in the write process. Involving only five drives instead of > >eight implies that the other three can be doing something else. > > Why would you ever need to write to _all_ of the drives, unless you are > writing an entire data-stripe's worth of data, in which case you'll be > writing to at least that many drives anyway? Never mind, I was thinking of parity regeneration. Too much other stuff going on in my head right now. I was thinking of the specific case where you need to read the other drives in order to calculate the parity. Assuming an initialized RAID, that's an invalid assumption, although you still have to read/write two blocks in that case. There's something else that's nagging at my memory... (I'm spending my day moving 8 million files, so disks are certainly the only thing on my mind right now) ... Joe ------------------------------------------------------------------------------- Joe Greco - Systems Administrator jgreco@ns.sol.net Solaria Public Access UNIX - Milwaukee, WI 414/342-4847 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199810280738.BAA10774>