X11SSM-F dressed up as a pig or no?

[John Doe]

i think I am searching for my perfect case since 4 years or so.... that one on the picture is the best what i could "find" within 4 years.

watercooling is absolutely no option for me. way too expensive, you need to maintain it and vibrations of the discs will multiply/add. so you will get problems of secondary and third vibrations (you know those which are overlaping adding peaks). probably noise might be another story.


thanks for that filter guy. need to check if delivery to europe is an option (and to what cost). need to check what 180mm is in that strange unit :p

 

[Bozon]

i think I am searching for my perfect case since 4 years or so.... that one on the picture is the best what i could "find" within 4 years.

watercooling is absolutely no option for me. way too expensive, you need to maintain it and vibrations of the discs will multiply/add. so you will get problems of secondary and third vibrations (you know those which are overlaping adding peaks). probably noise might be another story.


thanks for that filter guy. need to check if delivery to europe is an option (and to what cost). need to check what 180mm is in that strange unit :p

I was thinking any AC filter could be cut to size with a bread knife then reframed. so if shipping is the problem. I would try local hardware store. I like peanut M&Ms 180 would be a nice start. I’m not sure why that came up.:)

 

[elorimer]

Then there are the reports that temperature is not correlated with drive failure.

 

[Bozon]

Then there are the reports that temperature is not correlated with drive failure.

Yes when searching I came across an article that mentioned a google study about drive failure criteria. The study said heat wasn’t a factor. I could see google doing a study on how much they could abuse drives and have the drives still be reliable. It would mean 10’s of millions to their bottom line.

 

[Newfoundland.Republic]

Check out the BackBlaze drive error reports - there might be something there.

 

[Nakedape]

Then there are the reports that temperature is not correlated with drive failure.

Yes when searching I came across an article that mentioned a google study about drive failure criteria. The study said heat wasn’t a factor. [...]

I think both of these statements are false. If you look at the study, and specifically Figure 4 which I have reproduced below, there is in fact a clear negative correlation between temperature and AFR within the temperature range they had data on. Also of note is the sharp rise in AFR above 45C, which to me at least looks like a stark warning to anyone considering running their drives at elevated temperatures.

Source: https://static.googleusercontent.com/media/research.google.com/en//archive/disk_failures.pdf

 

[Bozon]

I think both of these statements are false. If you look at the study, and specifically Figure 4 which I have reproduced below, there is in fact a clear negative correlation between temperature and AFR within the temperature range they had data on. Also of note is the sharp rise in AFR above 45C, which to me at least looks like a stark warning to anyone considering running their drives at elevated temperatures.

View attachment 36252
Source: https://static.googleusercontent.com/media/research.google.com/en//archive/disk_failures.pdf

Thanks for finding the full article.

Here is the text describing the figure:

We first look at the correlation between average temperature during the observation period and failure. Figure 4 shows the distribution of drives with average temperature in increments of one degree and the corresponding annualized failure rates. The figure shows that failures do not increase when the average temperature increases. In fact, there is a clear trend showing that lower temperatures are associated with higher failure rates. Only at very high temperatures is there a slight reversal of this trend.

I believe that the * is the AFR (Average Failure Rate, it is the histograph over the bar chart ) So it looks like you should make sure that your drives aren't too cool, because below 35C the failure rate is largest, and only at temperatures above 45C do you see the graph start to got up. Between 36 and 45 look great. In fact, it looks like 46C shows the lowest AFR.

From their conclusion:

One of our key findings has been the lack of a consistent pattern of higher failure rates for higher temperature drives or for those drives at higher utilization levels. Such correlations have been repeatedly highlighted by previous studies, but we are unable to confirm them by observing our population. Although our data do not allow us to conclude that there is no such correlation, it provides strong evidence to suggest that other effects may be more prominent in affecting disk drive reliability in the context of a professionally managed data center deployment.

 

[John Doe]

I was thinking any AC filter could be cut to size with a bread knife then reframed. so if shipping is the problem. I would try local hardware store. I like peanut M&Ms 180 would be a nice start. I’m not sure why that came up.:)

i will check in the next days. A/Cs are not soo common in my place.

someone took a foot and defined it as a standard, someone else said, winter 1yyy was that and that cold, lets call it Fahrenheit.
so it wouldnt be too uncommon to give measurements in M&Ms you can stack haha :p

back to topic;
during my studies my prof showed us a lifespan of an LED.
it was basically a graph where you could see lifetime over head (measured on a nature constante, like kelvin :-D )

what you could see is, that the lifetime in the datasheets is always on room temperature (20°C).
that graph showed a drastic decreasing curve like an e function with every increase of heat.

And I think it is applicable for all semiconductor components. so it might not be the mechanic inside an HDD rather than all that ICs stuff on the PCB (just my opinion)

 

[Bozon]

I think both of these statements are false. If you look at the study, and specifically Figure 4 which I have reproduced below, there is in fact a clear negative correlation between temperature and AFR within the temperature range they had data on. Also of note is the sharp rise in AFR above 45C, which to me at least looks like a stark warning to anyone considering running their drives at elevated temperatures.

View attachment 36252
Source: https://static.googleusercontent.com/media/research.google.com/en//archive/disk_failures.pdf

But this is the opposite of what the graph shows, and the study concludes. The lowest failure rate happens at 46C ( 114.8F ), which is considered extremely hot for a drive. Most of the fan control scripts kick in to overdrive at 40C. The curve only really takes off at 51C ( 123.8F ). The graph really shows high failure rates climbing for drives that are relatively cool 30C ( 86F ) and below. Even their conclusion states this.

One of our key findings has been the lack of a consistent pattern of higher failure rates for higher temperature drives or for those drives at higher utilization levels. Such correlations have been repeatedly highlighted by previous studies, but we are unable to confirm them by observing our population. Although our data do not allow us to conclude that there is no such correlation, it provides strong evidence to suggest that other effects may be more prominent in affecting disk drive reliability in the context of a professionally managed data center deployment.

​

Pulling from this post:

Script to control fan speed in response to hard drive temperatures

The SuperMicro FAQ posted earlier just says it applies to X10 MBs. Silent on X9. So maybe out of luck? See here: http://www.supermicro.com/support/faqs/faq.cfm?faq=20882 I've tried it with a number of different X9, X10 and X11 boards. But the X9's and X10s were all E5 boards and the X11s...

www.ixsystems.com

​

Code from one of the scripts in this thread:

Code:

$max_allowed_temp = 38;    # celsius. you will hit 100% duty cycle when you HDs hit this temp.

Code from another of the scripts in this thread:

Code:

  # check for temperature greater than 39 deg C
  if ($temp > 39) { $case = 1; }
}

if ($case == 1) {
  # at least one hard drive is 40 deg C or higher
  # set fan speed control to Full
  `ipmitool raw 0x30 0x45 0x01 0x01`
} else {
  # all hard drive temperatures are 39 deg C or cooler
  # set fan speed control to Optimal
  `ipmitool raw 0x30 0x45 0x01 0x02`
}

 

[Bozon]

i will check in the next days. A/Cs are not soo common in my place.

someone took a foot and defined it as a standard, someone else said, winter 1yyy was that and that cold, lets call it Fahrenheit.
so it wouldnt be too uncommon to give measurements in M&Ms you can stack haha :p

back to topic;
during my studies my prof showed us a lifespan of an LED.
it was basically a graph where you could see lifetime over head (measured on a nature constante, like kelvin :-D )

what you could see is, that the lifetime in the datasheets is always on room temperature (20°C).
that graph showed a drastic decreasing curve like an e function with every increase of heat.

And I think it is applicable for all semiconductor components. so it might not be the mechanic inside an HDD rather than all that ICs stuff on the PCB (just my opinion)

Sure, I understand the vagaries of heat on semiconductors, but my point was that this study was done by Google, which probably has more hard drives than just about anyone on the planet, except maybe Amazon. Furthermore, they are motivated by cost to have the most comprehensive information on hard drive reliability as possible. Imagine, if the study had said drives fail very often above 30C, it would significantly impact the cost to run data centers, because of the extra cooling costs. But because the study says they can keep drives as hot as 45C without issue, they can save millions of dollars a year in cooling costs, and energy consumption. They just let the drives run at 45C with some method of ensuring that they stay at that level. I'm not quite prepared to toast marshmallows, on my future NAS, but now I can retort, but google says I can..., whenever someone says a drive needs to stay below 40C.

As an aside, this always makes me smile, when I think about units: The Smoot
Also, of course this one: Humorous Units of Measurements

 

[Nakedape]

But this is the opposite of what the graph shows, and the study concludes. The lowest failure rate happens at 46C ( 114.8F ), which is considered extremely hot for a drive. Most of the fan control scripts kick in to overdrive at 40C. The curve only really takes off at 51C ( 123.8F ). The graph really shows high failure rates climbing for drives that are relatively cool 30C ( 86F ) and below. Even their conclusion states this.

You said the study showed that heat wasn't a factor when it clearly does. Otherwise the AFR vs Temp in Figure 4 would be a constant horizontal line. I come from a different world where temperature ranges extend far beyond that of your typical data center environment, so I should have been clearer when I used the expression 'elevated temperatures'. What I meant by that was temperatures beyond 45C, say up to 65C which is what the WD Reds at least are rated for.

 

[Nakedape]

[...] but now I can retort, but google says I can..., whenever someone says a drive needs to stay below 40C.

Google did not say that you can, they said they could. In their environmentally controlled environment where relative humidity, temperature, vibration probably, etc all are carefully controlled. Unless you live in a Google datacenter, the results from this study should be extrapolated to your own setup with caution imo.

 

[Nakedape]

Bozon, thanks for those code snippets btw. I have much to learn and can't wait to get started, which reminded me that I need to continue hoarding parts for my box!

 

[Bozon]

Bozon, thanks for those code snippets btw. I have much to learn and can't wait to get started, which reminded me that I need to continue hoarding parts for my box!

Check the link I included in my post. Also the resources area, newer scripts may exist.