On Sun, Apr 8, 2012 at 7:40 AM, Jim Klimov <jimkli...@cos.ru> wrote:
> 2012-04-08 6:06, Richard Elling wrote:
>> You can't get past the age-old idiom: you get what you pay for.
> True... but it can be somewhat countered with DrHouse-age idiom:
> people lie, even if they don't mean to ;)

    I think both aspects over simplify the issue, but that is
generally true for idioms :-)

    I see Moore's Law in effect in a bunch of ways here, not all of them good.

    The available _capacity_ for a given _price_ at a given
_reliability_ and _performance_ increases with time. Two unintended
side effects of this increase in capacity _at_the_same_reliability_ is
that problems that a user was not likely to run into (statistically)
now occur on a more frequent basis and the time to correct (rebuild)
after failure is longer.

    It used to be that we looked at cost vs. capacity vs. performance,
the old adage was "pick two of the three". Today we need to add the
fourth parameter, reliability. I'm not sure the relationship between
the four is a linear one :-)

    So if you have <x> dollars (or euros or yen or ...) you need to
decide how much capacity / performance / reliability you can get for
that <x> and what you need. For my home server, where I am very cost
conscious, I went with Seagate SATA drives, but I went for the ES2
series which _are_ rated for 7x24 operation and have a (slightly)
better reliability spec than the lower cost drives I compared them
with. In the past, my need for an increase in capacity caused
replacement of hard drives. This time I'm not sure if I'll run out of
capacity before the drives reach end of practical service life and
start failing.

Paul Kraus
-> Senior Systems Architect, Garnet River ( http://www.garnetriver.com/ )
-> Sound Coordinator, Schenectady Light Opera Company (
http://www.sloctheater.org/ )
-> Technical Advisor, RPI Players
zfs-discuss mailing list

Reply via email to