Date: Sat, 16 Mar 2002 11:26:15 -0700 (MST) From: "M. Warner Losh" <imp@village.org> To: wkb@freebie.xs4all.nl Cc: bright@mu.org, mckusick@FreeBSD.org, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/coda coda_vnops.c src/sys/dev/ccd ccd.c src/sys/dev/md md.c src/sys/dev/vinum vinumdaemon.c vinuminterrupt.c vinumrequest.c vinumrevive.c src/sys/fs/hpfs hpfs_vnops.c src/sys/fs/msdosfs msdosfs_fat.c msdosfs_vnops.c ... Message-ID: <20020316.112615.113977724.imp@village.org> In-Reply-To: <20020316105245.A19669@freebie.xs4all.nl> References: <20020315185427.GK4857@elvis.mu.org> <200203152325.g2FNPwL61201@harmony.village.org> <20020316105245.A19669@freebie.xs4all.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <20020316105245.A19669@freebie.xs4all.nl>
Wilko Bulte <wkb@freebie.xs4all.nl> writes:
: On Fri, Mar 15, 2002 at 04:25:58PM -0700, Warner Losh wrote:
: > In message <20020315185427.GK4857@elvis.mu.org> Alfred Perlstein writes:
: > : In case no one has said it (which I doubt) you rule. :)
: >
: > daddr64_t is 9,444,732,965,739,290,427,392 bytes, or 9,444,732 PB (P
: > expressed in the SI units that disk makers use). This is >> the
: > largest disk arrays today (by a factor of 10^8). At a 20%/year growth
: > rate for disk sizes, that's 101 years before the largest disk arrays
: > get this big :-)
:
: FWIW: the biggest RFP I ever saw was a customer who wanted us to supply
: 1 PB in a single order. Which <clicketyclick> is approx 65 19" 42 racks
: at the moment. Soon to be half that number ;-)
Well, if the RFP is the biggest, then that reduces the life to only 88
years (since it is 10^7 times the size of the largest raid arrays
today :-). In 88 years, at the 20% growth rate, instead of 160Gish
drives being the high end, you'll have 1486000000G (or ~1500P) drives
at the high end.
Of course this does make the assumption that a moore's-like law of
only 20%/year does apply (which is a doubling every 46 months or so).
Over the last 10 years was a bad assumption. The largest drives in
early 1992 were about 2G, IIRC, and the largest drives in early 2002
are about 160G (someone will tell me of a bigger drive I'm sure :-), a
factor of 80 in 10 years, or 55%/year. Moore's law (double every 18
months) is 58.74%/year[*] (if bc is to be trusted). At 58.74%/year,
we're only 40 years away from this starting to be a problem. :-)
Warner
[*] bc -l
1.58740106^6
16.0000004857
(4 doubling in 6 years is one doubling every 1.5 years)
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020316.112615.113977724.imp>