Bit of a back story: we have had a number of servers over the years - probably 3 or 4 in total. In general - apart from the very first ‘SGI’ one I bought off Ebay (!) some 12+ years ago - they have been built and installed for us. Our current HP system is getting a bit tired. Its a quality piece of kit - cost us a bloody fortune! - but maintaining it is getting expensive. The drives are all SAS - small and expensive, but reliable.
Moving over to a NAS with SATA drives is a bit of a worry as two of our historic servers that were based on (allegedly ‘VERY’ good) SATA drives catastrophically failed…big time! Due to the nature of our business, the drives get hammered daily. I am hoping the WD Red drives are capable. There do seem to be a series of ‘datacentre quality’ WD drives that are a step up from the Red drives, but the Red are supposedly designed for RAID environments.
That’s a bit worrying - can I ask why you didn’t have an appropriate RAID setup then? Even if your primary drives fail you should never be in a situation like that where you have a catastrophic total failure.
The WD Reds are supposed to be excellent and the Blacks for normal use are great too.
It was advice we were given at the time. It was a RAID 5 setup. Unfortunately, for an unknown reason, two of the disks failed at the same time. RAID 5 is good for a single disk failure only. Its just the ‘blink’ way - if there is a rare event, it will occur to us!! Fortunately we had off-site backups, so loss was reduced.
Ouch! I guess it’s one of those live and learn things. At least you had off-site backups to fall back to. I remember working at a particular studio once upon a time who’s backup routine involved an automatic backup of everyone’s machines every evening… so long as they were turned on! So naturally lots of people in the office would boot down their machines in the evening, not knowing this would prevent backups. I remember at least 2 major panics for various people there as a result of corrupted live PSDs/max files and zero backups. It also, at one point, transpired that there were ZERO backups made for several weeks… I’m sure they changed their setup after that..!
What i think is if you have more then 2 people using nas, then you need a bigger nas like raid 10 or more, on top of that you can nest them so plug in one nas into another and mirror. This will ensure redundancy.
I got the nas with segate raid compliant drives, they are not your typical drives. I had bad experience sith WD though, not the raid ones but regular WD raptors that I had, all failed (I had like 6 drives) with in 1-3 years of life.
Lastly, larger facilities these days use isilons, they are expensive but I guess they are worth it
Create a radi 5-0 5 +1 .. etc … that is RELIABLE, require expensive and dedicated controllers that hambiente discs. Doing it with the controller program or motherboard, is CRAZY for heavy environments.
In our office we use Revodrive 1 tera Drives for file storage. In addition to a network FullDuplex teen Giga format. (A band of 100 to 100 to one side and the other, I mean two giga cable with separate directions.)
To TEEN driver you need a good card with a high cache. Over 512 MB integrated.
I like Synology. I have been using a 212, now with synology raid which is supposed to be redundant and allow you to expand by adding larger drives one at a time. Also have a usb3 drive attached to it for nightly back ups and a second drive for weekly backups.
Use several partitions to share as L: library (assets) n: administration, etc. And one I use for rendered frames. It was fairly easy to setup for a not too technical person and best of all the software updates have been pretty good.