Date: Wed, 14 Nov 2001 09:50:48 -0500 From: Bill Vermillion <bv@wjv.com> To: Carroll Kong <damascus@home.com> Cc: stable@FreeBSD.ORG Subject: Re: ATA hard drives Message-ID: <20011114095048.A14300@wjv.com> In-Reply-To: <5.1.0.14.2.20011114091338.03464ec0@netmail.home.com>; from damascus@home.com on Wed, Nov 14, 2001 at 09:19:30AM -0500 References: <bulk.37366.20011114002208@hub.freebsd.org> <bulk.37366.20011114002208@hub.freebsd.org> <20011114051132.A12369@wjv.com> <5.1.0.14.2.20011114091338.03464ec0@netmail.home.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Nov 14, 2001 at 09:19:30AM -0500, Carroll Kong thus spoke: > At 05:11 AM 11/14/01 -0500, Bill Vermillion wrote: > > > > Date: Tue, 13 Nov 2001 12:52:51 +0100 (CET) > > > From: Attila Nagy <bra@fsn.hu> > > > Subject: Re: ATA Hard Drives > > > > What is the largest/fastest ATA drives currently supported > > > > by -STABLE? > > > FreeBSD supports 48 bit addressing (introduced with ATA133 > > > standard), so the limit is quite theoretical. BTW, the largest > > > ATA drive I know of is the Maxtor 4G160J8 (160 GBs). Although > > > this may be the largest it isn't the fastest, because it spins > > > only with 5400 RPM. > >Just a comment on rotational speed. You should always check the > >manfacturers specs as the design of the head can greatly affect the > >speed of transfer to/from the platters. I have seen 5400 RPM > >drives move data faster than 7200 RPM drives. You don't see it > >often but RPM alone is not sole determining factor as much as > >legend would make you think it is. > Hm. I think it is more reasonable to say that density makes a big > difference. Grab an old 2.0 gig 7200 rpm micropolis scsi that is > not dead yet (those suckers died fast and hard), and plop in say > a new 40 gig maxtor 5400 rpm drive, and I am sure the maxtor will > be very close in performance. However, take any two modern drives, > the one with higher rpms is almost always going to win. Because, > at that point, density, seek times should be fairly close. Not neccesarily true. About 3-4 years ago I was researching drives and found the IBM 5400RPM SCSI drives were faster than the SEAGATE Baracudda 7200RPM drives. This was because the IBM's were using a new head design. Improved media can also mean that you can write bits just a 'bit' smaller. Typically this means a higher coercivity media so that the data bit is more finely focused. You can't compare old drives to new drives - but these were both current production drives. IBM does this now and then because they are doing a lot of head research and a lot of what we see in the high capacity drive markets today comes from them. They make nice little piles of money licensing things too. Over 50% of all advances in HD technology have come from Big Blue - and we are down to just handful of drive manufacturers in the world today. The best bet it to go the manufacturers site and look at the specs. The things to notice are the speed of data from the head to the buffer - listed as internal transfer rate. If the external rate is the same - eg Wide SE/LVD/Fibrechannel/etc - between the two drives then the one with the higher internal transfer rate will be the better performer. Bill 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?20011114095048.A14300>