March 21, 201610 yr I have 5 discs, 3 of which are HDD and 2 SDD, I use 3 VM with passthrough for vídeo. I use 2 of them for gaming and 1 for common access. I'd like to know the best way for optimizing the use of the discs. My thought was HDD(3TB parity), HDD(3TB disk 1), HDD(500GB disk2), 2 SSD(120GB CACHE1/2). Or to use the discs exclusively to the VM, HDD(3TB parity), HDD(3TB disk 1), HDD(500GB cache), 2SSD(120GB for VM1/VM2) and VM3 on cache or disk1.
March 21, 201610 yr I have 5 discs, 3 of which are HDD and 2 SDD, I use 3 VM with passthrough for vídeo. I use 2 of them for gaming and 1 for common access. I'd like to know the best way for optimizing the use of the discs. My thought was HDD(3TB parity), HDD(3TB disk 1), HDD(500GB disk2), 2 SSD(120GB CACHE1/2). Or to use the discs exclusively to the VM, HDD(3TB parity), HDD(3TB disk 1), HDD(500GB cache), 2SSD(120GB for VM1/VM2) and VM3 on cache or disk1. I personally like Option 1. It uses everything in the Official unRAID "supported" way (e.g. no disks running "outside of unRAID"), Virtualisation Files are stored on Cache etc. 3TB for Parity is a no brainer as your Parity drive HAS to be greater than or equal to the largest data disk in your array. It makes sense to have your 500GB disk in the Array too. With the SSD's having them in an BTRFS RAID-1 Array provides redundancy against disk failures BUT it is not the most efficient use of the space. You could convert the BTRFS Pool to RAID-0. This would mean your Cache Pool would have 240GB of space. As I mention above you are sacrificing redundancy for space, which I don't like BUT if you use mitigators (e.g. executing the mover more frequently - maybe every hour - AND backing up your VM's to the Array Daily) you are at least protected against data loss (assuming of course you have a backup strategy for your Array too) however IF there was an issue with one of the disks or the BTRFS filesystem got corrupt (YES it can happen and DOES - BTRFS is not the most stable FS yet) then your VM's would become unavailable until you restored service to your Cache. For your consideration.
March 21, 201610 yr As a side note IF you did want to run your Cache disks in BTRFS RAID-0 then in summary you would: - Create a DEFAULT RAID-1 Cache Pool - In Cache Settings (By Clicking on Cache 1 from Main Tab) scroll down to "Balance Status" and where you see ... -dconvert=raid1 -mconvert=raid1 replace it with -dconvert=raid0 -mconvert=raid1 and hit "Balance" Button! Note: IF you do this I would do this when there is NOTHING on the Cache disk(s) at all. Then if it goes wrong you have not been impacted at all. Reference reading: http://lime-technology.com/forum/index.php?topic=47408.msg454679#msg454679
March 21, 201610 yr Author As a side note IF you did want to run your Cache disks in BTRFS RAID-0 then in summary you would: - Create a DEFAULT RAID-1 Cache Pool - In Cache Settings (By Clicking on Cache 1 from Main Tab) scroll down to "Balance Status" and where you see ... -dconvert=raid1 -mconvert=raid1 replace it with -dconvert=raid0 -mconvert=raid1 and hit "Balance" Button! Note: IF you do this I would do this when there is NOTHING on the Cache disk(s) at all. Then if it goes wrong you have not been impacted at all. Reference reading: http://lime-technology.com/forum/index.php?topic=47408.msg454679#msg454679 OK thank you also thought about doing the passsthough of completely disk control , this would be a good option ? root@Frank:~# lspci | grep SATA 00:1f.2 SATA controller: Intel Corporation C600/X79 series chipset 6-Port SATA A HCI Controller (rev 05) 05:00.0 Serial Attached SCSI controller: Intel Corporation C602 chipset 4-Port S ATA Storage Control Unit (rev 05) 0a:00.0 SATA controller: Marvell Technology Group Ltd. 88SE9230 PCIe SATA 6Gb/s Controller (rev 10)
Archived
This topic is now archived and is closed to further replies.