On 9/24/07, Michael Cashwell <mboards@xxxxxxxxxxxx> wrote: > On Sep 24, 2007, at 11:53 AM, Greg Freemyer wrote: > > > On 9/24/07, Adil Mujeeb <mujeeb.adil@xxxxxxxxx> wrote: > > <snip> > >> adil@test:~> ls -lh test > >> -rw-r--r-- 1 adil group 2.0G 2007-09-24 23:51 test > >> adil@test:~> du -h test > >> 2.1G test <--------------------- why it is showing 2.1 GB > >> > >> I could not understand this difference. Could anybody explain this > >> behavioour? > > > > <snip> > > ie. "ls" is reporting Gibibytes and "du" is reporting Gigabytes. > > That does seem to be the case: > > 2147483648 / 1024^3 = 2.000 > 2147483648 / 1000^3 = 2.14748... Which rounds to 2.1. > > It's unfortunate that ls in this case doesn't use the new unit's > abbreviation. It does its math so as to return Gibibytes but then > fails to use the "GiB" unit so the user knows it did that. > > This whole issue really gets interesting when you start looking at > prices for data-center-class storage. You have big numbers for both > money and bytes. No one would think to say $1M is really $1,048,576 > but programmers and users disagree on whether 1MB is 1,048,576 or > 1,000,000 bytes. > > Fun! > > -Mike And it gets worse as data sizes grow. iirc: KB v. KiB = 2.4% error MB v. MiB = 4.8% error GB v. GiB = 7% error TB v. TiB = 11% error PB v. PiB = 15% error The disk drive manufactures have been using GB as 1,000,000,000 consistently for many years. It is the software teams that have not made the change (or are changing very slowly). Given that you can buy a TB size drive for under $500, it really is time for all this to get straightened out. An 11% unknown is just too much to ignore. Greg -- Greg Freemyer Litigation Triage Solutions Specialist http://www.linkedin.com/in/gregfreemyer The Norcross Group The Intersection of Evidence & Technology http://www.norcrossgroup.com -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
