Date: Thu, 16 Sep 2004 11:00:44 -0500 (EST) From: Sam <sah@softcardsystems.com> To: Jan Grant <Jan.Grant@bristol.ac.uk>, freebsd-hackers@freebsd.org Subject: Re: ZFS Message-ID: <Pine.LNX.4.60.0409161040480.28550@athena> In-Reply-To: <Pine.GSO.4.61.0409161528520.29724@mail.ilrt.bris.ac.uk> References: <41483C97.2030303@fer.hr> <Pine.LNX.4.60.0409151047230.21034@athena> <Pine.GSO.4.61.0409161010020.29724@mail.ilrt.bris.ac.uk> <Pine.GSO.4.61.0409161528520.29724@mail.ilrt.bris.ac.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 16 Sep 2004, Jan Grant wrote: > On Thu, 16 Sep 2004, Sam wrote: > >> Let's suppose you generate an exabyte of storage per year. Filling a 64-bit >> filesystem would take you approximately 8 million years. > > Hang on, I'm not sure I know where these numbers are coming from. > > 1PB is - what? 2^50 bytes? That looks closer to 2^64 than your > figures indicate. I'd imagine an exabyte a year ought to be topping out > after 16 years. I'm missing about half-a-dozen orders of magnitude > somewhere it seems. 1PB is indeed 2^50 bytes, but filesystems don't address on the byte, but on the block (1K, 4K, 8k, ...). The numbers I'm using assume the filesystem addresses on the sector, which is unrealistically small. Jack it up to a 16K blocksize and you jump a few hundred ZB in size. > Yes, it's a single filesystem. But the storage most likely won't be all > in one place. Making it look like it's accessible from one place is a > good thing. ... are you hinting at multiple globally remote block accessible storage sets? Otherwise I'm at a loss. Sam
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.60.0409161040480.28550>