Do you have a question? Post it now! No Registration Necessary. Now with pictures!
- Posted on
- OT: Servers, Many Discs vs Few?
June 16, 2011, 9:08 pm
rate this thread
because the NAS box offered the prospect of zero down time
if/when a disc failed.
But, with either solution, it takes quite a long time to fully
recover from a one or two-TB disc being replaced - although the
NAS box stays up and running during that time.
Right now, I'm running six 2-TB drives in the NAS box and a
hodge-podge of drives in the old WHS/backup box.
It's got me thinking that - despite the availability of bigger
and bigger hard drives - maybe a better solution
fault-tolerance-wise is a box with many smaller drives. e.g. a
bunch of 500 or even 250 TB drives.
Downsides would seem tb higher electric consumption and a higher
probability of any one drive failing.
Upside would seem tb faster recovery from a failed disc (and
therefore less exposure to the array getting hosed by a second or
third failure during recovery)
Does this sound reasonable?
Re: OT: Servers, Many Discs vs Few?
I think it's a tossup.
The question is - how often do you get a failure? I've had a network
storage with 2 1-TB drives for a couple of years now with zero failures.
It has crashed a couple of times due to power outages causing a
not-so-graceful shutdown, but recovers completely within 30-45 minutes
or so (yes, it's on a UPS but they don't last 5 days).
Remove the "x" from my email address
JDS Computer Training Corp.
Re: Servers, Many Discs vs Few?
As you add more drives the probablility of two concurrent drive failures
Raid 6 can survive two concurrent disk failures but is still limited by
the usual Raid drawbacks.
A Drobo can survive two concurrent disk failures, and it is also easily
expandable by adding more storage, and performance is not hampered
during a single drive failure.
- » ssh on command line: force using a group size (prime size) of 1024 (and no...
- — The site's Newest Thread. Posted in » Secure Shell Forum