SSD reliability in the real world according to google

[VirtualLarry]

You keep mentioning this exclusive burn-in process that Plextor does. Do you have any details about it or can we just safely assume they (stress) test the drives in a manner that ensures a low early failure rate?

http://www.goplextor.com/Product/Detail/M6_PRO#/Features


Tested to Extreme Standards

Every individual SSD undergoes Plextor's zero-failure zero-error standard during rigorous testing using the world's most advanced SSD testing facilities - FLEXSTAR® testing chamber.

100% Burn-in / Aging Test
48hrs Sustained Read and Write Test
4000 Times Idle Test – Sleep and Hibernation Test
250 Times Power Cycle – Cold and Warm Boot Test
1008hrs of Zero-Error Testing
Mean Time Between Failures of 2.4 Million Hours

 

[master_shake_]

it's a good thing ssd controller manufacturers use a RAID type method to help control errors and corruption on their drives then.

 

[tweakboy]

Milk it even more!

 

[frowertr]

The controller built onto the SSD won't find certain write errors like if the LBA map is/was incorrect during a write. You would only find out about that when actually attempting to read that data. This is when you would need to use external scrubbing/checksumming on every read via the filesytem.

RAID has nothing to do with fixing/correcting/identifying errors like this.

 

[coercitiv]

You keep mentioning this exclusive burn-in process that Plextor does. Do you have any details about it or can we just safely assume they (stress) test the drives in a manner that ensures a low early failure rate?

Every individual SSD undergoes Plextor's zero-failure zero-error standard during rigorous testing using the world's most advanced SSD testing facilities - FLEXSTAR® testing chamber.

100% Burn-in / Aging Test
48hrs Sustained Read and Write Test
4000 Times Idle Test – Sleep and Hibernation Test
250 Times Power Cycle – Cold and Warm Boot Test
1008hrs of Zero-Error Testing

Up until now, this is exactly the kind of testing that ensures low early failure rate. Where does it say they have no bad blocks in their NAND dies?

 

[Elixer]

http://www.goplextor.com/Product/Detail/M6_PRO#/Features


Tested to Extreme Standards

Every individual SSD undergoes Plextor's zero-failure zero-error standard during rigorous testing using the world's most advanced SSD testing facilities - FLEXSTAR® testing chamber.

100% Burn-in / Aging Test
48hrs Sustained Read and Write Test
4000 Times Idle Test – Sleep and Hibernation Test
250 Times Power Cycle – Cold and Warm Boot Test
1008hrs of Zero-Error Testing
Mean Time Between Failures of 2.4 Million Hours

So, they have 42 days of testing? They must have one huge facility to be able to do this...I wonder how much data they are writing out for all this testing as well?

 

[VirtualLarry]

Up until now, this is exactly the kind of testing that ensures low early failure rate. Where does it say they have no bad blocks in their NAND dies?

Well, what does "zero error" mean to you?

 

[coercitiv]

Well, what does "zero error" mean to you?

zero error = marketing term

Every individual SSD undergoes Plextor's zero-failure zero-error standard

So, they have 42 days of testing? They must have one huge facility to be able to do this...I wonder how much data they are writing out for all this testing as well?

Look at the way they present the information:

It sounds more like they do ~48hrs of stress testing, which they equate to 42 days of normal usage. Just like testing data retention is done with an accelerated method, keeping the drive at high temperatures in order to shorten data retention to more manageable levels.

 

[hojnikb]

Well, what does "zero error" mean to you?

No errors on host side. What ssd does internally really doesn't matter to end user, as long as data comes out error free.

 

[GoodEnough]

How about just tossing the SSD drive every 2 years?
Just a cost of doing business. Saves heartache.
I am going to buy another one right now.
Can someone suggest a good value 240GD SSD ?