Can I stop hard drives spinning up after resuming from sleep

[elsmandino]

Hi there,

I have got an HTPC/Server that has a lot of hard drives in it. The OS is run off a SSD and of all the hard drives, the one that really gets used is 1.5TB WD drive for recording of the television.

I have set Windows up so it turns off hard disks off after 20 minutes if they are not being used which is fine, but my problem is that my computer automatically goes into Sleep Mode (S3 hybrid sleep) when it is not recording anything and when it wakes up all the hard drives spin back up.

Is there any way to keep hard drives spun down even when resuming from sleep?

Thanks

 

[FliGi7]

No. Any drive that is connected to the system when it starts up or resumes from sleep is going to spin up. Windows attempts to recognize all connected drives when it does this and the only way to do that is to spin them up in order to access and interface with them. Either way, they're only spun up for 20 minutes because they're set to go to sleep after that, right? What is the issue with them doing this?

 

[elsmandino]

The thing that worries me is that I tend to record a lot whilst I am sleeping or at work, so the PC keeps waking up, recording for say half an hour and then going back to sleep. Isn't this a recipe for breaking hard drives?

I record a lot so they might be spun up and down six or seven times whilst I am not watching anything.

I suppose my other option is buy a Sata switch, so I can actually turn off hard drives that are not being used.

 

[FliGi7]

I think this is a topic of debate, whether spinning drives up/down often wears out a hard drive faster than just letting them remain spinning while not in use. I guess it would come down to proving whether spin up/down causes more damage to the bearings (like how most engine wear occurs at startup) or whether the act of the bearings take more wear by constantly being in use while the drive is spinning. Personally, I can't prove either and I'm not sure if anyone else can, empirically, prove that one or the other causes more harm more often. I would be interested if someone has some testing or data regarding this.

 

[elsmandino]

That is a really good point.

Thinking about if further I reckon that wear and tear aspect and the economics of spinning down (i.e. it takes masses more enery to spin up than to keep a disk spinning) are probably negligible.

The only important aspect is probably to do with heat/noise in your server in the first place.

 

[FliGi7]

That is a really good point.

Thinking about if further I reckon that wear and tear aspect and the economics of spinning down (i.e. it takes masses more enery to spin up than to keep a disk spinning) are probably negligible.

The only important aspect is probably to do with heat/noise in your server in the first place.

I would think lack of proper cooling might contribute more to premature death as well. However, I remember reading something a little while ago that even excessive cooling can kill them prematurely as disks are designed to operate within a certain temperature range. Of course, just because it's out there doesn't mean it's true, but it just goes to show how inexact of a science this really is. I'd say just do your best, within reason, and hope for the best (while making sure you're doing proper backups).