Sign in to follow this  
Followers 0
orion24

Reliability survey needs some rearrangement

11 posts in this topic

Looking eg at the WD listed models there are:

Raptor WD1500

Raptor WD360GD

Raptor WD740GD

VelociRaptor (4th Gen)

VelociRaptor (5th Gen)

VelociRaptor WD3000BLFS

Some are grouped in generations, others listed directly with the model. I'd suggest 1 raptor group with all the model numbers inside, or each generation in one group.

Then for the WD RE drives:

RE2 400 GB

RE2 500 GB

RE4

2 entire groups for the RE2, for 2 certain models. (I have 2 RE2 750GB, not listed). Again I would suggest 1 group for all the RE series.

WD could be like:

Caviar Green

Caviar Blue

Caviar Black

Caviar Red

Raid Edition

Raptor

Mobile Green

Mobile Blue

Mobile Black

SAS

AV

Archive

or like (based on WD site):

Desktop

Raptor

Mobile

NAS

Enterprise

Audio/Video

If you guys have a lot of work, we could be of some help if you need it.

Share this post


Link to post
Share on other sites

I think it needs a lot more complexity than that, especially since drives should be listed by specific model as the platter densities change to make the survey at all remotely useful.

Grouped in generic categories like "WD Red, WD Blue, WD Black" tells nothing about reliability of specific platforms within the general family, and is quite useless.

Share this post


Link to post
Share on other sites

I think it needs a lot more complexity than that, especially since drives should be listed by specific model as the platter densities change to make the survey at all remotely useful.

Grouped in generic categories like "WD Red, WD Blue, WD Black" tells nothing about reliability of specific platforms within the general family, and is quite useless.

Maybe so, with WD. Seagate is pretty clear with the a product number defining a specific drive. This should be the case with WD as well. Wherever there are situations with different platter sizes and the same product numbers, there could be different options. Eg WD10EZEX (500GB platters), WD10EZEX (1TB platters).

Share this post


Link to post
Share on other sites

Unfortunately we have no real idea what the reliability is on any of the drives other than poor in many cases. MTBF has always been a theoretical joke and actual reported failure rates vs. true failure rates leaves a lot of error so it's difficult to get any accurate sense of reliability though certain models may have higher than typical failure rates, such as SATA drives in the past five years or so.

The situation with SSDs is even worse than HDDs when you consider both compatibility and reliability issues. There are firmware upgrades required on some SSDs just to get them to function at all in certain PCs.

It's really quite a mess of rushed-to-market crap at this point.

Share this post


Link to post
Share on other sites

My pipe-dream is that end-users could just upload the output of smartctl to your reliablity database. There is already well-maintained FOS code for parsing the output in the gsmartcontrol gui-wrapper. You could attract uploads by making a website that accepts a smartctl.txt upload and rewards you with a gui-report, similar to what gsmartcontrol (and other websites) already do.

Something like this was suggested in 2010:

http://forums.storagereview.com/index.php/topic/28805-hd203wi-puis-power-up-in-standby-support/page__p__262736__hl__%2Bsmartctl+%2Bsubmit__fromsearch__1#entry262736

And Brian said (Posted 24 July 2010 - 08:48 AM):

Not a bad idea...we can probably work that into the reliability database when we re-work it.

If you are willing to host the backend database, I'm sure the open-source community could easily come up with the web-page front-end!

Share this post


Link to post
Share on other sites

That is, the idea of entering data into the website via smartctl output would be that the database could then create entries based on the make, model, and firmware number as self-reported by each drive. That might make it easier to find a particular drive and/or drive-family. (Or would this require a complete re-work of your database?).

Share this post


Link to post
Share on other sites

I removed all my submissions, because the database isn't useful anymore in the current state and maintaining them serves no purpose. Not only are most current drive models missing, there are just not enough submissions for recent drives to lead to useful statistical data.

It's great that you can tell how well a Seagate Barracuda ATA IV survived in the stone age, but for most recent drives you have only a dozen comments and sometimes you even have zero submissions.

I think that's the main reason, why the promised revamp of the reliability database is never gonna happen.

Share this post


Link to post
Share on other sites

It would take a large volume of data/submissions to obtain any reasonably accurate indication of reliability/compatibility on HDDs or SSDs. I know of no such database and it would require a lot of effort to develop and maintain one that is meaningful, which is likely why they don't exist.

I would strongly recommend that those interested in purchasing an HDD or SSD perform due diligence as the compatibility and reliability of SSDs in particular is borderline fraud IMO. I'd also suggest going to the HDD/SSD maker's website and reading the actual product warranty as there are quite a few surprises these days as storage products become commodities. If your data and uninterrupted PC use is important, then you will want to know what to expect as you may very well be using the warranty on many currently available storage products.

Also note that advertised SSD performance figures typically do not apply to SSDs installed in Apple PCs, which is often not disclosed until after you install the product and see dismal performance in comparison to the SSD advertised performance.

Share this post


Link to post
Share on other sites

A reliability survey database for SSDs would have been useful in the past five years, but it wasn't there. A missed opportunity.

Share this post


Link to post
Share on other sites

A reliability survey database for SSDs would have been useful in the past five years, but it wasn't there. A missed opportunity.

Since there is no practical means to create an accurate SSD reliability database the point is moot.

Share this post


Link to post
Share on other sites

The accuracy of the database has always been questionable since the time we took over the site. We do want to provide meaningful data here and are continually talking to vendors about possible ways to accomplish this.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0