RAID issues on Western Digital Hard disk

Not sure if there is a specific problem the dealer was referring to, but it should at least function in a proper array. Since they do use some variable spin technology to reduce power consumption, it may cause some slow downs in the array, but I can't say for sure how they perform from experience.

There's nothing wrong whatsoever with using a green drive in a RAID array. You just have to understand why you're doing it and if it makes the best sense for you.

RAID arrays are used for two main purposes:

- improving performance by spreading the I/O load over multiple drives (much like a multi-core CPU). This is what RAID-0 does for you.

- improving reliability by creating redundant copies of the data. This is what RAID-1 does (by keeping two complete copies of the data), and also what RAID-5 does (by keeping parity information). (Both these RAID levels also give you improvements in read performance by spreading the I/O load over multiple spindles, but they hurt write performance - only slightly for RAID-1 and a lot for RAID-5).

The only reason a green drive might be an issue in a RAID array is if you're trying to use RAID to get better performance. In that case, since green drives have generally poorer performance it would make more sense to choose a higher-performing drive (such as the WD "black" series).

But if you have some green drives sitting around already, then you'll get better performance out of them by putting them into a RAID-0 set. It just won't be as good as the performance you'd get out of drives that perform better in the first place.

I prefer performance over data security so i should go for RAID 0 right?

My dealer is ready to exchange the HDD so shall i implement RAID on 2 HDD of western Digital caviar green 640GB or 2 HDD of Seagate 500GB

which one is better for RAID seagate or WD caviar green.

WD's "RE" (RAID Edition) HDDs support Time-Limited Error Recovery ("TLER"):

http://www.wdc.com/en/products/productcatalog.asp?langu...


As a non-TLER HDD fills up with data, the error detection firmware
might take too long, and the RAID controller may drop that HDD
from a RAID array.


MRFS

That's right.
Well you didn't say specifically which Seagate drives you're referring to (and I don't have any particular familiarity with Seagate drives anyway), but basically you want to get a drive with the fastest access times you can find. For example, a drive with 9ms access time is better than one with 11ms access time.

Interesting. Do you know what the timeout period would be before a typical RAID controller would drop the drive? That could be a real issue with green drives since they basically power themselves down after a few seconds of inactivity and require a few seconds to get back into "ready" status.

One would hope that the timeout period could be customized to deal with that sort of issue.

Typically around 8 seconds with integrated driver/host RAID solution. Some hardware RAID cards allows you to change that value. Mine is through LSI's MegaCLI which I have set to 30s as I'm not using HDDs with TLER in a array.
Of course with any WD drives you can use a utility downloadable from their site (forgot what it's called) to enabled TLER which is defaulted to 6 seconds timeout I think.

I am pretty sure no drive spins itself down without a command given to it.

On the GP drives TLER is disabled by default, on the "Raid Edition" drives, this feature is enabled by default.

In either case, if there is a bad sector on one of the RAID members; it will get kicked out of the array. TLER can't protect against that, but it can protect against a stalling/hanging system while one RAID-member disk is performing recovery.

From the Western Digital website:

Western Digital manufactures desktop edition hard drives and RAID Edition hard drives. Each type of hard drive is designed to work specifically in either a desktop computer environment or a demanding enterprise environment.

If you install and use a desktop edition hard drive connected to a RAID controller, the drive may not work correctly unless jointly qualified by an enterprise OEM. This is caused by the normal error recovery procedure that a desktop edition hard drive uses.

When an error is found on a desktop edition hard drive, the drive will enter into a deep recovery cycle to attempt to repair the error, recover the data from the problematic area, and then reallocate a dedicated area to replace the problematic area. This process can take up to 2 minutes depending on the severity of the issue. Most RAID controllers allow a very short amount of time for a hard drive to recover from an error. If a hard drive takes too long to complete this process, the drive will be dropped from the RAID array. Most RAID controllers allow from 7 to 15 seconds for error recovery before dropping a hard drive from an array. Western Digital does not recommend installing desktop edition hard drives in an enterprise environment (on a RAID controller).

Western Digital RAID edition hard drives have a feature called TLER (Time Limited Error Recovery) which stops the hard drive from entering into a deep recovery cycle. The hard drive will only spend 7 seconds to attempt to recover. This means that the hard drive will not be dropped from a RAID array. Though TLER is designed for RAID environments, it is fully compatible and will not be detrimental when used in non-RAID environments.

There they have a point, in an enterprise environment. You don't want your box being down/unresponsive for 2 minutes because one of many disks is doing recovery.

But for home users, the result is the same: after the x seconds with TLER enabled, the device will give up its recovery and report a read failure. The RAID-engine will then kick the drive out of the array. Some try again for a few times (the same sector) but they will still kick the drive out.

On Linux/BSD you can see timeouts happening, and the only issue i ran across was setting disks to spin down to conserve power in a RAID-config would lead to timeouts and disks being kicked out. After setting the timeout value to 60 seconds, this issue disappeared. FreeBSD 8 is also known to automatically increase the timeout of the device when its spun down.

So if you're using Linux or BSD, even a 2 minute recovery time wouldn't be disastrous if you configured it right. It could cause the server to hang when any disk is doing recovery, but for home users getting it to work is more important than 2 minutes downtime. Where enterprise environments use redundancy and would want to swap out any disk as soon as possible whenever the smallest suspicion of malfunction arises.

So, TLER is not a "must-have" in home RAID environments, and may even make otherwise recoverable data irrecoverable, unless you use redundancy like RAID1/3/4/5/6. Either way, having an operating system that allows you to set the timeout to a high value would be more valuable than having TLER, for home users.

ok if western digital has this issue what about seagate 500GB sata 16mb cache (Model No. : ST3500418AS)

i am confused with western digital caviar green 640 GB 16mb cache and seagate 500GB (Model No. : ST3500418AS). so which one u prefer for raid 0.

Pick the WD Green drives if you'll be storing large files like when downloading stuff from the internet. Even a music file of 3MB is considered large. Windows has many files of less than 4 kilobytes, which demands different performance characteristics.

So the Green drives are excellent for large files (sequential transfers), they may be less suitable to load your operating system, applications and games on. It won't run terribly slow, but it won't be as fast as other modern non-Green drives.

The green drives do keep very cool since they use little power and are also very quiet, so i would recommend them.

The green drives do not work with RAID. I had RAID 10 (mirrored and striped) and I lost EVERYTHING, programs and data. Before I lost everything I had failures about 3 times before the final crash. For a high end machine that cost over $2000, I will spend 5 times of that for my time getting back to speed and that does not include the data lost, or lost billings. I had 4 tetrabyte drives. According to my geek, even the blacks are not recommended by Western Digital with RAID, only the servers.

With the greens, they go to sleep and cant wake up fast enough for RAID. WD does recommend their very expensive server drives, and now I have 2 two tetrabyte server drives from Seagate, mirrored, but not striped (sounds like the original JB). WD black drives may fair better but they are not that reliable either, so if you want RAID, get the server drives.

So to summarize: The issue with WD GREEN drives is they can be too slow to respond within normal RAID timeouts; however you can make them work if you change your raid software/controller to allow up to 30 seconds for them to respond. Your raid will run slower than if you had the twice-as-expensive WD raid drives, but it will work.

Does that summary sound about right?

I hope so as I just bought two green 2B drives and I'm going to put them in a Linux box as a raid1. I'm not going to boot off it, and I don't mind a long pause before my files start streaming if I can get a cheap raid at half the price.

WD RE 7200 RPM hard drives are engineered to thrive in a high duty cycle RAID system
while still offering traditional low cost-per-capacity desktop value. Designed and
manufactured to server-class standards, WD RE drives provide RAID-specific,
time-limited error recovery (TLER) to significantly reduce drive fallout caused by extended
drive error-recovery processes and are tested to 1 million hours MTBF. These drives
include an 8 MB (SB) or 16 MB (YS) cache for faster time to data and fluid dynamic
bearings (FDB) to reduce heat, vibration, and noise.
Drives need to be Raid "re" or server drives. When running a raid o stripe, the TLER times out because drive size doubles, and the drive drops out of the array, causing corruption of the drive in many cases. make sure the drive has been approved for raid or server drive, This dep cycle error checking software isnot in a the raid drives, and you will not have any furtherproblems with raid failures

Hi
I know this is too late , but i just wonder is TLER is required when i do a software raid like windows raid . as i got WD20EARS which is TLER not supported so i changed from my intel RAID MOBO to Windows Software Raid 5 hoping that TLER is not reuired in SOFT Raid so my HD will not dropped if they take much time to repair the Bad Sectors
what do you think?

This topic has been closed by Mousemonkey