 |
 |
|
 |
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
On Tue, 10 Nov 2009 10:12:57 +0000, Invisible wrote:
>> There's
>> nothing, though, that says you can't create a mirrored set, let the
>> drives mirror, and then "break the mirror" and take one drive offline.
>> I've known people who have done that and used that for disaster
>> recovery when upgrading systems.
>
> This is a very, very dumb way to do backup. A file-level copy will be
> drastically faster. (It doesn't involve mirroring all the useless empty
> sectors.) IME, mirroring a disk typically takes something like 10 hours,
> regardless of capacity. (Lower-capacity drives are usually
> correspondingly slower too.)
>
> Also, if you do a file-level copy, you have options such as compressing
> the data and putting multiple backups on a single backup harddisk, doing
> differential or otherwise partial backups, and so forth.
Depends on what you're trying to accomplish. If the purpose is fast
recovery, mirroring is a very good way to ensure you have a workable
system if you swap the drives out.
Doing a file-level copy might be faster, but then you get to go in and
muck about with permissions afterwards. Mirroring takes care of all of
that for you.
Remember - I've got decates of working with server technology behind me
here. ;-)
>>> An often-encountered backup strategy is to copy everything onto an
>>> external USB HD and then put that somewhere. I'm not sure that all
>>> this turning the drive on and off won't just wear it out faster.
>>
>> See my anecdotal evidence in reply to Stefan. Two identical units, one
>> powered on and off regularly, one that was left on 24x7. Guess which
>> one failed? Not the one that has been turned on and off regularly for
>> 5 years now. The one that was plugged in and running for 3 years
>> solid.
>
> In theory, until the disks are spinning at full speed, you don't get
> that "cushion of air" for the heads to "fly" on, which should result in
> wear. Of course, no doubt manufacturers know all about this and have
> come up with ways to at least reduce the problem...
I would expect they do. I used to know a guy who worked for Quantum
years ago (I wonder what he's up to these days), and he said the
engineering was quite impressive.
Jim
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
On Tue, 10 Nov 2009 10:17:16 +0000, Invisible wrote:
>>> I don't know, man... Backing up spinning disk to... spinning disk? Is
>>> that such a sensible idea?
>>
>> Why not? It's a backup. It's not like you can read a tape without
>> spinning it. What are you going to back it up to, FLASH RAM?
>
> Tape operates at much lower speeds. And since the only people who use
> tape are people who want seriously reliable backup storage, it tends to
> be very well engineered. (And stupidly expensive...)
Never used DAT drives for backup, have you? ;-) Horrible quality of
storage media, and terrible shelf life IME.
Jim
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
On Tue, 10 Nov 2009 07:38:30 +0000, Orchid XP v8 wrote:
> Wanna take a guess what the customary place to install programs is?
> Yeah, it's C:\Program Files.
Then there's also the button that lets you choose a different location to
install to - part of pretty much every modern installation program on
Windows.
Jim
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
On Tue, 10 Nov 2009 10:21:30 +0000, Invisible wrote:
>>> ...yeah, shopping online tends to be cheaper. Quality is another
>>> matter, of course.
>>
>> Buy from reputable dealers, and quality isn't an issue.
>
> I don't mean to imply that quality is *necessarily* an issue, just that
> it *can* be more of an issue. As you say, it depends who you buy from.
Exactly. I started using newegg because system builders I know who build
for small businesses used them. I rarely go to a place I've never heard
of and bought something. Oddly, the svideo cable I mentioned before I
did go to a relatively unknown company - but I didn't drop more than
about 10 bucks on the cable, either, and when I do something like this, I
usually start with something cheap. I did this with a place that sold
through Amazon - bought a $3 iPod battery replacement kit - that worked,
so ordered a laptop battery from them for my stepson's laptop. Both have
worked well. :-)
> I often buy from Maplin because 1) I don't have to wait for it to be
> delivered, 2) their staff actually have a frickin clue, and 3) they
> don't argue about returns.
Sometimes expediency is worth the extra cost - as is the ability to get
something replaced without a hassle.
Jim
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
> In theory, until the disks are spinning at full speed, you don't get that
> "cushion of air" for the heads to "fly" on, which should result in wear.
> Of course, no doubt manufacturers know all about this and have come up
> with ways to at least reduce the problem...
There's probably just a reserved portion of the platter that is used for
spin-up/down operations. The only tricky bit is to get the head over to
that portion when the power is suddenly cut.
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
>> Tape operates at much lower speeds. And since the only people who use
>> tape are people who want seriously reliable backup storage, it tends to
>> be very well engineered. (And stupidly expensive...)
>
> Never used DAT drives for backup, have you? ;-)
No, only for the last 6 years.
> Horrible quality of storage media, and terrible shelf life IME.
Really?
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
>> I often buy from Maplin because 1) I don't have to wait for it to be
>> delivered, 2) their staff actually have a frickin clue, and 3) they
>> don't argue about returns.
>
> Sometimes expediency is worth the extra cost - as is the ability to get
> something replaced without a hassle.
Maplin are pretty much the best electronics shop around.
I mean, apart from being the *only* place in all of Christendom which
can sell you (say) a 50kohm resistor, their staff have a clue about what
they're supposed to be selling, and if you take something back they will
exchange it or give you your money back (provided it isn't obviously
busted).
Their website utterly sucks, by the way...
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
>> In theory, until the disks are spinning at full speed, you don't get
>> that "cushion of air" for the heads to "fly" on, which should result
>> in wear. Of course, no doubt manufacturers know all about this and
>> have come up with ways to at least reduce the problem...
>
> There's probably just a reserved portion of the platter that is used for
> spin-up/down operations.
I'm 98% certain this is the case. However, that still means that this
part of the disk wears down, and also that the heads themselves wear down.
> The only tricky bit is to get the head over to that portion when the
> power is suddenly cut.
This is probably fairly trivial. In the old days you used to have to run
the "park" command. But today... well, have you ever noticed that
"click" noise when you unplug the power? I imagine that's the magnet in
the voice coils parking the heads. ;-)
In fact, I've disassembled a HD and observed that the arm strongly
resists any attempt to remove it from the home position. It keeps
wanting to click back into place, with the heads on the outer edge of
the platters.
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
>> Wanna take a guess what the customary place to install programs is?
>> Yeah, it's C:\Program Files.
>
> Then there's also the button that lets you choose a different location to
> install to - part of pretty much every modern installation program on
> Windows.
There is *usually* such a button, yes. Not always though.
And it's not unheard-of for a program to malfunction if you change this
setting during installation. (You'd think if they offered you the
option, they'd implement it so it works. But not always...)
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"scott" <sco### [at] scott com> wrote:
> > In theory, until the disks are spinning at full speed, you don't get that
> > "cushion of air" for the heads to "fly" on, which should result in wear.
> > Of course, no doubt manufacturers know all about this and have come up
> > with ways to at least reduce the problem...
>
> There's probably just a reserved portion of the platter that is used for
> spin-up/down operations. The only tricky bit is to get the head over to
> that portion when the power is suddenly cut.
I think they don't park the heads on the platter any more, they're moved out of
the platter radius before spindown.
Moving the head out of the way in case of power failure is not so problematic
I'd think. The positioning coils of the heads are constantly under current and
cutting this current will create a force. You "just" have to make it so that a
power failure exerts a force in the head mechanism that whips the head into
park. ;o)
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|
 |