1 in 10^14 vs 10 in 10^15

[tarui]

I was reading up about the differences between the WD red and red pro and saw the difference in error rates

wd red error rate 1 in 10^14 (~10tb)
wd red pro error rate 10 in 10^15(~100tb)

am I right to say they are not the same because in the latter, one will encounter 10 errors but it's likely they encounter it at say, after the 30TB mark (which is greater than the 10tb trigger point of 1 in 10^14)?

 

[anodos]

Error rates per spec sheets appear to be the same.
http://www.wdc.com/wdproducts/library/SpecSheet/ENG/2879-800022.pdf
http://www.wdc.com/wdproducts/library/SpecSheet/ENG/2879-800002.pdf

The real question is whether you want or need 7200RPM drives (as opposed to 5400RPM ones).

 

[Bidule0hm]

If you use ZFS then the UBER isn't a concern because the errors will be corrected by ZFS ;)

 

[GrumpyBear]

If you are planning on lots of random access or high IOPS then the Red Pro or Hitachi Disks make sense. With just Gigabit Ethernet the standard consumer NAS disks will be more than adequate to saturate the link and produce less heat.

But yes, a magnitude of 10 increase in the Uncorrected Bit Error Rate does mean a disk should have errors less frequently.

 

[cyberjock]

If you use ZFS then the UBER isn't a concern because the errors will be corrected by ZFS ;)

Assuming you have redundancy that can recover from the error. ;)

 

[Bidule0hm]

Of course ;)

I didn't specify it because why to use ZFS without redundancy?

 

[Ericloewe]

I was reading up about the differences between the WD red and red pro and saw the difference in error rates

wd red error rate 1 in 10^14 (~10tb)
wd red pro error rate 10 in 10^15(~100tb)

am I right to say they are not the same because in the latter, one will encounter 10 errors but it's likely they encounter it at say, after the 30TB mark (which is greater than the 10tb trigger point of 1 in 10^14)?

I disagree with that interpretation.

Beyond infant mortality, accidents and mechanical failure, you can use a uniform distribution to approximate the error rate. 1/10^14 = 10/10^15 Both numbers are the exact same thing.

I'm sure someone at marketing got a raise for realizing they could trick people by using a larger exponent with a larger numerator to represent the exact same number.

 

[anodos]

I disagree with that interpretation.

Beyond infant mortality, accidents and mechanical failure, you can use a uniform distribution to approximate the error rate. 1/10^14 = 10/10^15 Both numbers are the exact same thing.

I'm sure someone at marketing got a raise for realizing they could trick people by using a larger exponent with a larger numerator to represent the exact same number.

The official spec sheets show 1/10^14 for both drive types. My guess is the specs were written by engineers, and the other materials by marketeers. I did notice that the max operating temp for the red pro is lower, which makes me wonder whether it's statistically more likely to fail at higher temps than the WD red.

 

[Bidule0hm]

I did notice that the max operating temp for the red pro is lower, which makes me wonder whether it's statistically more likely to fail at higher temps than the WD red.

I think the red pro is more or less the red and they just lowered the max temp in the datasheet so they can rise the MTBF, the warranty, ... without any changes to the physical design.

 

[marbus90]

I think the red pro is more or less the red and they just lowered the max temp in the datasheet so they can rise the MTBF, the warranty, ... without any changes to the physical design.

Red without pro = Green (cheap consumer stuff) with NAS firmware
Red Pro = Black, 7200rpm, big cache, amazingly fast with NAS firmware

so there is a whole lot of changes to the physical design. it's 5400rpm vs. 7200rpm.

 

[Bidule0hm]

Actually it doesn't change what I think, I was wrong on the model, replace red not pro by black in my post. But it's only my personal thought ;)

 

[anodos]

Red without pro = Green (cheap consumer stuff) with NAS firmware
Red Pro = Black, 7200rpm, big cache, amazingly fast with NAS firmware

so there is a whole lot of changes to the physical design. it's 5400rpm vs. 7200rpm.

Actually, they have the same amount of cache. See here: http://promotions.newegg.com/wd/14-...NA&gclid=CJirjeGc6MQCFQrqfgodllgAHQ&gclsrc=ds

 

[marbus90]

Actually, they have the same amount of cache. See here: http://promotions.newegg.com/wd/14-...NA&gclid=CJirjeGc6MQCFQrqfgodllgAHQ&gclsrc=ds

That's just some rambling from back in the days (2011). :)