Date: Fri, 30 Oct 1998 08:19:52 -0800 (PST) From: Tom <tom@uniserve.com> To: "Mikhail A. Sokolov" <mishania@demos.net> Cc: Rich Winkel <rich@chumbly.math.missouri.edu>, stable@FreeBSD.ORG Subject: Re: RAID support in FBSD? Message-ID: <Pine.BSF.4.02A.9810300816090.6748-100000@shell.uniserve.ca> In-Reply-To: <19981030104833.55472@demos.su>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 30 Oct 1998, Mikhail A. Sokolov wrote: > On Thu, Oct 29, 1998 at 01:35:02PM -0800, Tom wrote: > # On Thu, 29 Oct 1998, Rich Winkel wrote: > # DPT has a whole line of host based RAID controllers. They are probably > # the best you can get. Very wide operating system support, so they are > ^^^^^^^^^^^^^^^^^^^ Usually, DPT is __very__ slow to operate in degraded > mode, which makes it way unusable when such things happen. Since > the RAID (5, for instance) are supposed to be redundant/working even when > in degraded mode, you want SCSI-SCSI box, something from www.infortrend.com. All RAID5 units are slow in degraded mode. That is the whole nature of RAID5. I've had DPT PM334 cards go into degraded mode, and no one even noticed, except for the people next to the machine room heard the DPT alarm. I don't know of any host based DPT cards better than DPT. You may find a SCSI-SCSI faster, but that isn't a host based adapater. > Let me as well issue the following phrase not proving it, since it'd need > more than 2 pages of advocacy: since you will not really achieve anything > more than 2mb/sec (values provided for ultra1-wide) in production environment, > you really want SCSI-SCSI RAIDs which are much more easy to operate, > more redundant (see Tom's definitions/disadvantages list). > > I neither work for infortrend, nor dpt, of course. Tom Systems Support Uniserve To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.02A.9810300816090.6748-100000>