Nas

Ruidoso Bill

Senior Member
Just added a buffalo 8TB Terasation to my network for back up, what would folks familiar with Raid to suggest my best option for Raid setup. The Terastation has 4 X 2Tb hard drives.
 

PapaST

Senior Member
Very nice. I definitely would check which RAIDS it will support. I'm guessing you'll be using RAID5. For instance my FreeNAS that I recently build supports the normal RAID0,1,5 etc. But also can run some ZFS RAID options.

RAID5 should yield you 6TB of space.
 

Ruidoso Bill

Senior Member
Very nice. I definitely would check which RAIDS it will support. I'm guessing you'll be using RAID5. For instance my FreeNAS that I recently build supports the normal RAID0,1,5 etc. But also can run some ZFS RAID options.

RAID5 should yield you 6TB of space.

They seem pretty expensive new but thanks be to ebay, 350 plus shipping, 4 2tb drives is almost that much. The thing looks bullet proof and heavy and it does support Raid 5 so will probably go that route.
 

PapaST

Senior Member
RAID 6 will give you 2 drive fault tolerance but will likely yield less available space. RAID 10 I'm not too familiar with... typically the higher RAID levels are hybrids of lower ones. But with 4 disks it is worth looking into. Some people don't like RAID 5 because of the single disk redundancy.

What are you going to store on the NAS?
 

Ruidoso Bill

Senior Member
Pictures and web sites I've built for clients, I already have it all on individual 1 TB extternal drives and some on cloud but the idea is redundant back up. I read Raid i0 is actually 1+0. I hard drive failure seems enough with Raid 5.
 
Last edited:

PapaST

Senior Member
Sounds like you have it all covered. Check the manual for alerts. I'm willing to bet it has some type of alert system where it will send you an email or alert you in some way if there is a drive failure or predictive failure.
 

Ruidoso Bill

Senior Member
Sounds like you have it all covered. Check the manual for alerts. I'm willing to bet it has some type of alert system where it will send you an email or alert you in some way if there is a drive failure or predictive failure.

Thanks for the help and advice, appreciated. At first it seems like giving up 2 tb of storage is rather expensive but if redundancy is the goal not too bad. I notice it also has 2 usb ports so gotta read and see if I can hang on more.
 
Last edited:

Steve B

Senior Member
Bill, For a 4 drive array your best choices are RAID-5 (3D+1P) or RAID-10 (2D+2D). RAID-5 will give you more space, RAID-10 (depending on how it is implemented) can give you a little more reliability and better read speeds. With good quality drives, and alerts set up to let you know if a drive fails, RAID-5 is usually the best overall choice.
 

randyspann

Senior Member
Just added a buffalo 8TB Terasation to my network for back up, what would folks familiar with Raid to suggest my best option for Raid setup. The Terastation has 4 X 2Tb hard drives.
Hi Bill - I have the same Buffalo NAS as yours at work (TS3400D87E). I set it up raid 5. I think it is very good bang for the buck. I'm an IT/Systems guy and other options are much more expensive. I haven't found the throughput to be so great, but for your application, you will find it more than adequate. Four 2 TB drives will get you 6 TB usable (roughly) (5491.6 GB actual because the drives are 1831.4 GB usable).

Are you accessing it from Windows?

By the way, I just notice a firmware update is available (version 1.0.1)
 
Last edited:

kluisi

Senior Member
Bill, For a 4 drive array your best choices are RAID-5 (3D+1P) or RAID-10 (2D+2D). RAID-5 will give you more space, RAID-10 (depending on how it is implemented) can give you a little more reliability and better read speeds. With good quality drives, and alerts set up to let you know if a drive fails, RAID-5 is usually the best overall choice.

I thought there were a couple of problems with RAID 5 implementations. The one I was thinking of was the Write Hole:

"Write hole" phenomenon in RAID5, RAID6, RAID1, and other arrays.

When I searched for it, I also found this information which I was not aware of before, but it seems to be a new issue that is popping up with the increasing size of disks. It seems that if one drive in a RAID 5 array fails and is replaced, the error rate for huge disks is still low enough that when it tries to read all of the parity information off of the other disks in the array, it is likely that one of the reads for the parity bits will fail and cause your entire rebuild of the array to fail with it:

Why RAID 5 stops working in 2009 | ZDNet

Yikes.
 

Steve B

Senior Member
The rebuild times for arrays of large capacity drives can be extremely long. Two main problems. 1. The drives that are normally used in these less expensive arrays are usually pretty slow. If you notice, a lot of times they don't even give a rotation speed for them. 2. Most of these drives were also never designed to be running 100% of the time which means one of the most likely times for a drive to fail is during a long rebuild. In short, they are great choices for large capacity near-line storage but don't keep all of you eggs in one basket. Either use them as a backup or back them up somewhere else. RAID protection isn't 100% failure proof.
 

Steve B

Senior Member
I thought there were a couple of problems with RAID 5 implementations. The one I was thinking of was the Write Hole:

"Write hole" phenomenon in RAID5, RAID6, RAID1, and other arrays.

When I searched for it, I also found this information which I was not aware of before, but it seems to be a new issue that is popping up with the increasing size of disks. It seems that if one drive in a RAID 5 array fails and is replaced, the error rate for huge disks is still low enough that when it tries to read all of the parity information off of the other disks in the array, it is likely that one of the reads for the parity bits will fail and cause your entire rebuild of the array to fail with it:

Why RAID 5 stops working in 2009 | ZDNet

Yikes.
A very good reason to always have your NAS device attached to a UPS. The write hole issue normally occurs due to a power outage during a write. My four drive Netgear NAS is protected by a UPS that can keep it running for at least 20 minutes. More than enough time to shut it down gracefully.
 

randyspann

Senior Member
All true! But ... history, I've been running raid 5 on 96 plus servers for 14 years. I have never had an incident causing loss of data. Yes array rebuild times after a drive replacement can be slow, but this is probably not a real active system: read/write performance doesn't have to be high. Hence the low cost.
 

Ruidoso Bill

Senior Member
Hi Bill - I have the same Buffalo NAS as yours at work (TS3400D87E). I set it up raid 5. I think it is very good bang for the buck. I'm an IT/Systems guy and other options are much more expensive. I haven't found the throughput to be so great, but for your application, you will find it more than adequate. Four 2 TB drives will get you 6 TB usable (roughly) (5491.6 GB actual because the drives are 1831.4 GB usable).

Are you accessing it from Windows?

By the way, I just notice a firmware update is available (version 1.0.1)

I went with Raid 5, still close to 6 TB which isn't too bad. I did see the firmware update when I was setting it up and it is now 1.59 if I recall. Moved it into my network closet so now it's out of the way. Do you leave yours powered up or turn it on when needed? I am thinking since mechanical wear is what ruins hard drive by only powering it on for use then shut it down would extend life, thoughts?
 

randyspann

Senior Member
I went with Raid 5, still close to 6 TB which isn't too bad. I did see the firmware update when I was setting it up and it is now 1.59 if I recall. Moved it into my network closet so now it's out of the way. Do you leave yours powered up or turn it on when needed? I am thinking since mechanical wear is what ruins hard drive by only powering it on for use then shut it down would extend life, thoughts?

Always on. Harddrives will fail maybe 1 out of 20 in 3-4 years, the rest will outlive the computer/NAS they are installed in. A backup system has much more value if it is always up! Contrary to a previous post to this thread, all harddrives are of very good quality (the problem is usually in the controller, usb drives). (My opinion)
 

DraganDL

Senior Member
I went with Raid 5, still close to 6 TB which isn't too bad. I did see the firmware update when I was setting it up and it is now 1.59 if I recall. Moved it into my network closet so now it's out of the way. Do you leave yours powered up or turn it on when needed? I am thinking since mechanical wear is what ruins hard drive by only powering it on for use then shut it down would extend life, thoughts?

"since mechanical wear is what ruins hard drive by only powering it on for use then shut it down would extend life". True. Try to use them only when needed. All of them have their predicted MTBF (lifespan) expressed in number of hours while being turned on. Anyone concerned with the lifespan of their HDDs should consider buying a "server" disk(s), which can be expected to last more than 10 years, even if constantly "on": http://www.hgst.com/hard-drives/enterprise-hard-drives/enterprise-sata-drives/ultrastar-7k3000
 
Last edited:

Steve B

Senior Member
History - I work for one of the companies that make large scale enterprise arrays. We are talking over 1000 drives in a single array. Drives fail. Sometimes at the worst possible time. I have a Buffalo Terastation that is probably about 10 years old. I don't use it much anymore because I got a NetGear ReadyNAS NVX that has better features than the older TeraStation. I had a drive fail in my TeraStation within two years after I purchased it. Not all drives are created equal. If I had it to do again I would buy one of the empty enclosures and add four drives purchased separately so I knew exactly what I was getting. Some hard drive manufactures no longer use the MTBF (Mean Time Between Failures) as a measure of reliability. Seagate now uses AFR (Annualized Failure Rate). Hard disk drive reliability and MTBF / AFR The problem is that finding this type of information, as well as rotational speed, is difficult for many of the desktop class of drives. This information is usually readily available for Enterprise class drives.
 
Top