I'm building a new system in an 8 bay server and am considering buying three 8tb HGST drives + parity. I can't find anything that explains how big the parity drive needs to be? I'm thinking it needs to be 8tb, but could it be split across two smaller / less expensive 4tb drives at a slower rpm since parity is only written on a sync?
Trying to do this right the first time but I can't find anything on the FAQ or manual pages that explain proper parity sizing? Thanks
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
with latest version of SnapRAID v11, split parity is supported. so you can use 38TB for data and 24TB for parity. The parity drive (or sum of splitted parity drives) must be as big as the largest data drive. Usually, the parity drive get's a little bit for filled than data driver due to alignment of parity block.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I want to use 10Tb disks and I have been thinking is it possible to partition the drives to have part data and the other part parity ?
ie) if I had 8x 10TB disks 75% Data and 25% parity could this still work as a redundancy system for a disk failure ?
also what would be best running 2x 10tb snapraid parity drives or having the parity file stored on
6x3tb ZFS Raidz-2 file system ?
or
1x 10tb snapraid parity drive and 10tb parity file on 5x 3tb ZFS Raidz-1 ?
Last edit: Master CATZ 2017-04-03
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
This is the way I see it. People set up RAID systems for safekeeping data. I set up my systems to be as basic and simple as possible. That way, if a drive dies, any drive, the recovery is as simple as possible. No worrying about if this drive dies, what happened the spanned drive. If a drive dies in my system, I replace that drive, recover that drive, and I'm back up and running. Simple with no dependencies. If needed, I can take that data drive out and work with it directly.
Of course, other people are welcome to set up their RAID systems as they see fit. But I'm a big believer in making things as simple as possible. Especially when recovering data.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The main reason I am using SnapRaid is its easier to scale up in size for my Multimedia collection than it is scaling up an ZFS system ,
I was just thinking if the parity file was stored on soemthing with data redundancy and striped for performace it might work better ( plus I could make use of the 3tb drives I never used )
plus I have had a multiple drive failure with ZFS that would have caused me to loose everything before
( had a bad batch of 3tb disks that did not even last 1 week ! )
thankfully I still had my original 2tb array in storage
bright side of file of RAID over File System's is you don't lose everything
and I beilve this software can even dig around to tell you what you have lost ?
I know how Snapraid works with 1 parity drive but how does it work with extra ones ? are the extra's just a clone of the first ?
as I have noticed you can switch between 1 drive and multiple drives with out any Parity calculations?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I know how Snapraid works with 1 parity drive but how does it work with extra ones ? are the extra's just a clone of the first ?
No, the second parity is completely different.
as I have noticed you can switch between 1 drive and multiple drives with out any Parity calculations?
No, you can only downgrade to fewer paritiy levels instantly. Adding more levels requires snapraid reads the enire array in order to calculate the added parity.
plus I could make use of the 3tb drives I never used
You can still do that. You can use 4x3 TB as a 1x12 TB parity disk.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm building a new system in an 8 bay server and am considering buying three 8tb HGST drives + parity. I can't find anything that explains how big the parity drive needs to be? I'm thinking it needs to be 8tb, but could it be split across two smaller / less expensive 4tb drives at a slower rpm since parity is only written on a sync?
Trying to do this right the first time but I can't find anything on the FAQ or manual pages that explain proper parity sizing? Thanks
Largest drive for parity.
with latest version of SnapRAID v11, split parity is supported. so you can use 38TB for data and 24TB for parity. The parity drive (or sum of splitted parity drives) must be as big as the largest data drive. Usually, the parity drive get's a little bit for filled than data driver due to alignment of parity block.
I am in the same boat with the split parity
I want to use 10Tb disks and I have been thinking is it possible to partition the drives to have part data and the other part parity ?
ie) if I had 8x 10TB disks 75% Data and 25% parity could this still work as a redundancy system for a disk failure ?
also what would be best running 2x 10tb snapraid parity drives or having the parity file stored on
6x3tb ZFS Raidz-2 file system ?
or
1x 10tb snapraid parity drive and 10tb parity file on 5x 3tb ZFS Raidz-1 ?
Last edit: Master CATZ 2017-04-03
This is the way I see it. People set up RAID systems for safekeeping data. I set up my systems to be as basic and simple as possible. That way, if a drive dies, any drive, the recovery is as simple as possible. No worrying about if this drive dies, what happened the spanned drive. If a drive dies in my system, I replace that drive, recover that drive, and I'm back up and running. Simple with no dependencies. If needed, I can take that data drive out and work with it directly.
Of course, other people are welcome to set up their RAID systems as they see fit. But I'm a big believer in making things as simple as possible. Especially when recovering data.
True
The main reason I am using SnapRaid is its easier to scale up in size for my Multimedia collection than it is scaling up an ZFS system ,
I was just thinking if the parity file was stored on soemthing with data redundancy and striped for performace it might work better ( plus I could make use of the 3tb drives I never used )
plus I have had a multiple drive failure with ZFS that would have caused me to loose everything before
( had a bad batch of 3tb disks that did not even last 1 week ! )
thankfully I still had my original 2tb array in storage
bright side of file of RAID over File System's is you don't lose everything
and I beilve this software can even dig around to tell you what you have lost ?
I know how Snapraid works with 1 parity drive but how does it work with extra ones ? are the extra's just a clone of the first ?
as I have noticed you can switch between 1 drive and multiple drives with out any Parity calculations?
No, the second parity is completely different.
No, you can only downgrade to fewer paritiy levels instantly. Adding more levels requires snapraid reads the enire array in order to calculate the added parity.
You can still do that. You can use 4x3 TB as a 1x12 TB parity disk.