Loading ...
Sorry, an error occurred while loading the content.

Re: [nslu2-linux] Re: Which drives do you recommend?

Expand Messages
  • David Given
    ... Hash: SHA1 pezmannen_77 wrote: [...] ... Not necessarily --- there s some evidence to suggest that most of the wear happens during spin-up and spin-down
    Message 1 of 16 , May 2, 2007
    • 0 Attachment
      -----BEGIN PGP SIGNED MESSAGE-----
      Hash: SHA1

      pezmannen_77 wrote:
      [...]
      > Isn't that exactly what you want your drives to do? Spinning them down
      > will make them live longer, wouldn't it?

      Not necessarily --- there's some evidence to suggest that most of the wear
      happens during spin-up and spin-down (maximum torque). Spinning down mainly
      makes them use less power.

      - --
      ┌── dg@cowlark.com ─── http://www.cowlark.com ───────────────────
      │ "This is the captain. We have a little problem with our reentry sequence,
      │ so we may experience some slight turbulence and then explode." --- Mal
      │ Reynolds, _Serenity_
      -----BEGIN PGP SIGNATURE-----
      Version: GnuPG v1.4.6 (GNU/Linux)
      Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

      iD8DBQFGOLfhf9E0noFvlzgRAscRAKCqQThP1uFKNEO7wwHiCzomiq/LWwCgqgj7
      WLsjutCIaAG9UWntm+IOK/s=
      =1jgP
      -----END PGP SIGNATURE-----
    • Brian Wood
      ... There is a lot of argument over that issue. Personally I prefer to let them spin. It will wear the bearings more, but is kinder to the electronics. Plus, I
      Message 2 of 16 , May 2, 2007
      • 0 Attachment
        David Given wrote:
        >
        >
        > pezmannen_77 wrote:
        > [...]
        >> Isn't that exactly what you want your drives to do? Spinning them down
        >> will make them live longer, wouldn't it?
        >
        > Not necessarily --- there's some evidence to suggest that most of the wear
        > happens during spin-up and spin-down (maximum torque). Spinning down mainly
        > makes them use less power.
        >

        There is a lot of argument over that issue. Personally I prefer to let
        them spin. It will wear the bearings more, but is kinder to the electronics.

        Plus, I do not like the delay waiting for the drive to spin up when I go
        to access my NFS-mounted drives. The drives are accessed enough that the
        spun-down time would be much less than the spun-up time anyway.

        Essentially it's the "lightbulb effect". How many lightbulbs have you
        seen burn out just as they are turned on, as opposed to ones that are
        just sitting there burning away and decide to fail.

        It's a bad analogy perhaps, since the electronics of the drive is not
        powered down. This is why the arguments persist.

        The best thing you can do to increase drive life is to keep them as cool
        as you can.

        But the bottom line is the decision should be up to the user, and not be
        forced on him or her by hardware limitations.
      • Chris Watts
        Hi I wouldn’t advise a Maxtor drive either, mine just gave up the ghost and reading reviews this seems common across the Maxtor range! L Cheer Chris Hi I
        Message 3 of 16 , May 2, 2007
        • 0 Attachment

          Hi

           

          I wouldn’t advise a Maxtor drive either, mine just gave up the ghost and reading reviews this seems common across the Maxtor range! L

           

          Cheer

          Chris

        • Drew Gibson
          ... Some folks with a lot of experience with hard disks might beg to differ about cooling... http://labs.google.com/papers/disk_failures.html They are dealing
          Message 4 of 16 , May 2, 2007
          • 0 Attachment
            Brian Wood wrote:
            David Given wrote:
              
            pezmannen_77 wrote:
            [...]
                
            Isn't that exactly what you want your drives to do? Spinning them down
            will make them live longer, wouldn't it?
                  
            Not necessarily --- there's some evidence to suggest that most of the wear
            happens during spin-up and spin-down (maximum torque). Spinning down mainly
            makes them use less power.
                
            There is a lot of argument over that issue. Personally I prefer to let
            them spin. It will wear the bearings more, but is kinder to the electronics.
            
            Plus, I do not like the delay waiting for the drive to spin up when I go
            to access my NFS-mounted drives. The drives are accessed enough that the
            spun-down time would be much less than the spun-up time anyway.
            
            Essentially it's the "lightbulb effect". How many lightbulbs have you
            seen burn out just as they are turned on, as opposed to ones that are
            just sitting there burning away and decide to fail.
            
            It's a bad analogy perhaps, since the electronics of the drive is not
            powered down. This is why the arguments persist.
            
            The best thing you can do to increase drive life is to keep them as cool
            as you can.
            
              
            Some folks with a lot of experience with hard disks might beg to differ about cooling...
            http://labs.google.com/papers/disk_failures.html
            They are dealing with regular 3.5" disks which like to run constantly, not 2.5" laptop drives which _are_ engineered to be spun up and down frequently.
            Google's conclusions:-
            1. Hard drives prefer to be hot rather than cool (over cooling causes more failures than under cooling, unless the drives are >5 yrs old.)
            2. Load is not a factor (so thrash 'em drives!)

            But the bottom line is the decision should be up to the user, and not be
            forced on him or her by hardware limitations.
              
            I like to let mine run, spin them down to save battery life or the planet only.

            regards,

            Drew
          • Brian Wood
            ... I got three of the 500GB Maxtor units very cheap on sale. I have removed the PATA drives from 2 of them for internal use in other machines. They were both
            Message 5 of 16 , May 2, 2007
            • 0 Attachment
              Chris Watts wrote:
              >
              >
              > Hi
              >
              >
              >
              > I wouldn’t advise a Maxtor drive either, mine just gave up the ghost
              > and reading reviews this seems common across the Maxtor range! L

              I got three of the 500GB Maxtor units very cheap on sale. I have removed
              the PATA drives from 2 of them for internal use in other machines. They
              were both Seagate drives, their "best" consumer perpendicular-type units.

              Perhaps it is the USB case that "gave up the ghost"? Ive had no problems
              with the drives, but they are in properly-cooled cases now, not those
              strange things with the psychedelic lights on the front.
            • Brian Wood
              ... At some point, if your spun-down time is short enough, you are actually damaging the planet more by spinning down as the startup surge might exceed the
              Message 6 of 16 , May 2, 2007
              • 0 Attachment
                Drew Gibson wrote:
                >>
                > I like to let mine run, spin them down to save battery life or the
                > planet only.

                At some point, if your spun-down time is short enough, you are actually
                damaging the planet more by spinning down as the startup surge might
                exceed the savings.
              • Philip Pemberton
                ... Seconded - I ve got a pair of the 400GB drives in a RAID1 config - they re very nice. Well, except for the fact that you need to hit the power switch to
                Message 7 of 16 , May 2, 2007
                • 0 Attachment
                  MrE wrote:
                  > I'm using a Seagate Pushbutton Backup 350 GB drive, which is *very*
                  > quiet and spins down by itself; I didn't need to make any config changes
                  > to get it to spin down.

                  Seconded - I've got a pair of the 400GB drives in a RAID1 config - they're
                  very nice. Well, except for the fact that you need to hit the power switch to
                  turn 'em on if the power fails. If I can find a way to get them apart without
                  breaking them, I'll install a power-on mod of some kind in them. Warranty,
                  what warranty? :P

                  --
                  Phil. | (\_/) This is Bunny. Copy and paste Bunny
                  ygroups@... | (='.'=) into your signature to help him gain
                  http://www.philpem.me.uk/ | (")_(") world domination.
                • Chris Watts
                  ... Perhaps the Maxtor drives were so unrealiable that Seagate replaced them with their own drives but retained the Maxtor name on the usb case? Just don t
                  Message 8 of 16 , May 3, 2007
                  • 0 Attachment
                    >Brian Wood wrote:
                    >
                    >
                    >I got three of the 500GB Maxtor units very cheap on sale. I have removed
                    >the PATA drives from 2 of them for internal use in other machines. They
                    >were both Seagate drives, their "best" consumer perpendicular-type >units.

                    >Perhaps it is the USB case that "gave up the ghost"? Ive had no problems
                    >with the drives, but they are in properly-cooled cases now, not those
                    >strange things with the psychedelic lights on the front.

                    Perhaps the Maxtor drives were so unrealiable that Seagate replaced them with their own drives but retained the Maxtor name on the usb case? Just don't buy a old one off ebay or make sure you don't get old stock that been sitting on a self for ages then!
                  • Brian Wood
                    ... That s the problem with buying external USB units, you never really know what s inside. Usually you can t even tell if it is a SATA or a PATA drive, but
                    Message 9 of 16 , May 3, 2007
                    • 0 Attachment
                      Chris Watts wrote:
                      >
                      >
                      >>Brian Wood wrote:
                      >>
                      >>
                      >>I got three of the 500GB Maxtor units very cheap on sale. I have removed
                      >>the PATA drives from 2 of them for internal use in other machines. They
                      >>were both Seagate drives, their "best" consumer perpendicular-type >units.
                      >
                      >>Perhaps it is the USB case that "gave up the ghost"? Ive had no problems
                      >>with the drives, but they are in properly-cooled cases now, not those
                      >>strange things with the psychedelic lights on the front.
                      >
                      > Perhaps the Maxtor drives were so unrealiable that Seagate replaced them
                      > with their own drives but retained the Maxtor name on the usb case? Just
                      > don't buy a old one off ebay or make sure you don't get old stock that
                      > been sitting on a self for ages then!


                      That's the problem with buying external USB units, you never really know
                      what's inside. Usually you can't even tell if it is a SATA or a PATA
                      drive, but that matters only if you're going to rip it out I guess.

                      I've often seen external units selling for less than a same-sized bare
                      drive. I think the reason is the makers can throw whatever they sweep up
                      off the floor into those cheap USB cases and the buyers will be happy as
                      long as the capacity is close to what's on the box. Issues like speed,
                      latency etc. are all made academic by the USB interface.

                      Also, the market seems to want SATA units these days, and the external
                      boxes are a way to get rid of excess PATA drives.

                      I tried to buy a 20 or 40 GB the other day, all I really needed was 2GB
                      to hold an OS. Of course the smallest I could easily find was 160GB.

                      Of course if you want to buy a *real* hard drive (Ultra-320 SCSI at
                      15,000 RPM) you still pay through the nose. Quality always costs, but is
                      usually worth it in the end.
                    • mattvillago
                      ... i like seagate because most of their products have a 5 year warranty.
                      Message 10 of 16 , May 3, 2007
                      • 0 Attachment
                        --- In nslu2-linux@yahoogroups.com, "pezmannen_77" <pezmannen_77@...>
                        wrote:
                        >
                        > Hi
                        >
                        > I'm gonna buy the nslu2 next week. First i decided to by a MyBook
                        > 500gb from WD, but reading the Wiki and here has made be reconsider.
                        > Looks to me that nobody has managed to spindown these drives?
                        >

                        i like seagate because most of their products have a 5 year warranty.
                      • Phil Endecott
                        ... That s true, but only just. Here are some numbers from a Seagate datasheet; I suspect that other drives are similar: Idle: 7.2 W Peak: 33.6 W How long
                        Message 11 of 16 , May 3, 2007
                        • 0 Attachment
                          > At some point, if your spun-down time is short enough, you are actually
                          > damaging the planet more by spinning down as the startup surge might
                          > exceed the savings.

                          That's true, but only just. Here are some numbers from a Seagate
                          datasheet; I suspect that other drives are similar:

                          Idle: 7.2 W
                          Peak: 33.6 W

                          How long does it take to spin up? Let's say 3 seconds. 3 x 33.6 is
                          near enough 100 J, which is about what it takes during about 14 seconds
                          idling. So you save power overall as long as your average period of
                          time spun down is greater than about 14 seconds. Surely no-one would
                          set a spin-down time short enough to see that sort of average behaviour.

                          The Seagate datasheet also quotes "50,000 contact start-stops" in its
                          "reliability" section; I don't really know what that means, but maybe
                          it means spin down/up cycles. If you want the drive to last 5 years of
                          24/7 operation, that's about one start-stop per hour.

                          If I were still using disks (which I'm not - no moving parts here
                          except the keyboard and the printer) I would probably choose a spin
                          down period of the order of 10 to 30 minutes.


                          Phil.
                        Your message has been successfully submitted and would be delivered to recipients shortly.