Pooling Clouddrives

[kalk]

Hi,

 

I was wondering if you could help me look over the correct settings for my pool. I'm using the latest beta of all products, including Scanner.

 

I want to mirror two virtual drives (drive A & drive B ) which are offsite providers, and have additional duplication on a subset of files using a cloud drive on my NAS (drive C).

 

I want writes to the drive to land on the system drive first (which is an SSD) and the NAS second (in case space runs out on the system drive), and automatic balancing to take place during the night. After that, I'd like only the folders with 3x duplication to go on the NAS.

 

What confuses me is the options under Balancing, as I don't think I fully understand the Duplicated/Unduplicated checkboxes.

 

Do I add the system drive to the pool, using the SSD optimizer plugin, or let the Clouddrive expandable cache handle it? Do I designate the NAS a SSD since it's much faster than the offsite virtual drives?

 

Thanks in advance!

[Christopher (Drashna)]

Do I add the system drive to the pool, using the SSD optimizer plugin, or let the Clouddrive expandable cache handle it? Do I designate the NAS a SSD since it's much faster than the offsite virtual drives?

 

For reference, the SSD Optimizer balancer does clear out the "SSD" drive.  So if you're okay with it being emptied, then this should be fine.

 

 

Otherwise, there isn't a good way to do this. 

 

However, new data is/will be written to the cache drives before uploading. 

 

 

 

 

What confuses me is the options under Balancing, as I don't think I fully understand the Duplicated/Unduplicated checkboxes.

 

For the "Drive Usage Limiter" balancer, right? 

If so, ....

Unduplicated data is that, data that is not duplicated.

As for duplicated data, we do differentiate the data as "original" and "duplicate". It's all the same. So if you have only one drive with "duplicated data" enabled for, it will run into issues. 

[kalk]

OK, I think I'll skip the extra step of adding the system drive to the pool and rely on Clouddrive's caching.

 

 

For the "Drive Usage Limiter" balancer, right? 

If so, ....

Unduplicated data is that, data that is not duplicated.

As for duplicated data, we do differentiate the data as "original" and "duplicate". It's all the same. So if you have only one drive with "duplicated data" enabled for, it will run into issues. 

 

Yes, that's right.

 

So in my case, drive A and B the (offsite cloud providers) should have "duplicated" checked, and drive C (NAS) should have both "unduplicated" and "duplicated" checked, is that correct? Does that mean that new data will be added to drive C first, since I have disabled "Real-time duplication"?

 

What about the "Duplication Space Optimizer", should that be used in my scenario? 

[kalk]

I forgot to mention that drives A & B are the same size, while drive C is smaller. 

[Christopher (Drashna)]

If you have real time duplication disabled, then yes, I believe that is what you'd see.  

 

However, if that isn't happening, then try using the "SSD Optimizer" balancer plugin, and set the C: drive as the "SSD".

Note: you don't have to have SSDs, this balancer essentially creates a write cache, but anymore, a majority of people are using SSDs for this.

 

https://stablebit.com/DrivePool/Plugins