View Full Version : how you fare with harddisks ?
Huupi
April 15th, 2008, 04:48 AM
its a widespread understanding that the higher temperatures,the fail rate of harddrives will equally go up,but from google there is an interesting survey on harddrive failures.
http://www.tgdaily.com/content/view/30990/113/
and also interesting http://luminous-landscape.com/forum/index.php?showtopic=24568
Hairy Coo
April 15th, 2008, 04:57 AM
Thanks Huupi-interesting!
{QUOTE-> Only drives with an age of six months or younger show a decidedly higher probability of failure when put into a high activity environment. Once the drive survives its first months, the probability of failure due to high usage decreases in year 1, 2, 3 and 4 - <-QUOTE}
That seems the danger period
{QUOTE-> However, drives with scan errors turned out to be ten times more likely to fail than drives without scan errors <-QUOTE}
This also confirms my experience-had a six months old drive WITH scan errors-sure enough was faulty:wacko:
If you are lucky,the impending failure will give plenty of notice(scan errors),so back it up.
Seagate exchange was amazing,from Sydney to Indonesia then Thailand and back within days-physically and communications.
EASTER
April 15th, 2008, 05:10 AM
I have one Maxtor Drive that runs too warm for my taste and i'm taking measures soon to transfer it's data because in my opinion HEAT can spell deterioration or even buid up pressure toward other circuitry in the motherboard and wiring.
I have noticed the many other drives i have been going strong for 3-4-5 years etc. and seem to be doing just fine so far.
Brian K
April 15th, 2008, 05:41 AM
The Google study is interesting in that cooler drives failed more often in the first 3 years than drives at higher temperatures. Contrary to our gut feeling.
markymoo
April 15th, 2008, 07:24 AM
Yes it not mentioned well but you should insist on keeping your hard drives cool it increases reliability. Thats why it good to have a front intake fan blowing over the drives keeps then cooler or wherever they be. I can fry an egg on top of my Raptors with no fan around but with a fan they feel cool to the touch. The casing act like one big heatsink. The long sides of the drives can be the hottest. A couple of these help http://us.st11.yimg.com/us.st.yimg.com/I/svcompucycle_1995_118980311
ErikAlbert
April 15th, 2008, 08:16 AM
I have two internal harddisks, one external harddisk and quality CD/DVD's as second backup for objects, I don't want to lose.
Although it is in theory possible, that 3 harddisks fail at the very same time, this is as good as impossible in practice. So I'm not worried.
Almost everything was broke in my old computer, except my IBM-harddisk, which was 7 years old.
Huupi
April 15th, 2008, 11:02 AM
{QUOTE-> Yes it not mentioned well but you should insist on keeping your hard drives cool it increases reliability. Thats why it good to have a front intake fan blowing over the drives keeps then cooler or wherever they be. I can fry an egg on top of my Raptors with no fan around but with a fan they feel cool to the touch. The casing act like one big heatsink. The long sides of the drives can be the hottest. A couple of these help http://us.st11.yimg.com/us.st.yimg.com/I/svcompucycle_1995_118980311 <-QUOTE}
Yes,Against all odds(seemingly),harddrives have a longer livespan with higher temp.which seems contradict the myth in which drives last longer in lower temperatures.Its weird,its also against the laws of physics !
Message from the google boys is that temperature is NOT the common denominator,but one of the many that influenced livespan. pffft.
markymoo
April 15th, 2008, 12:02 PM
Maybe it was a disgruntled employer, no can't be everybody happy who works at Google lol. Whats the common factor on drives that fail after 2-3 years? They been running hot for a long time and they been turned on an off alot. What kills parts? Heat. Alot hard drives are kept in cases without giving consideration to there heat. It could be drives had a slight knock even before you bought it and it detoriated quicker.
{QUOTE-> Although this is often claimed by manufacturers of HDD cooling products, the temperature does not directly affect the performance of the drive in any significant way - seek times and data transfer rate will stay the same, no matter how hot the drive is running. Old hard drives had a thermal recalibration loop - from time to time, the hard drive would make a short pause and recalibrate itself. When the drive warmed up, the platters expand, and their geometry changes, thus making recalibration necessary. With such drives, good cooling resulted in less frequent recalibration, and therefore did improve performance, especially for Audio/Video applications. However, these days are long over - all recent drives with embedded servo do not have a thermal recalibration loop. Temperature will not affect performance in any way.
<-QUOTE}
{QUOTE-> The platters (disks) carry the data. They are made from light aluminum or optical glass and ceramic, and are coated with a magnetic layer. When the drive heats up, the platters expand, changing their size and geometry. This destroys the magnetic layer on the surface where the data is stored, and creates defects and micro cracks. <-QUOTE}
Who knows they could be possible myths or not. Heating up and cooling down of those controller chips must shorten there life.
Huupi
April 15th, 2008, 12:41 PM
Yes that was something not to be argued in their conclusion,the temperature changes and more important temperature differences and their impact on the mechanical parts.IMO this is the foremost cause in failing.
One more could be a heavy fragmented drive as it is obvious that the mechanical parts are in for excessive stress.
So solution is keep it cool or warm and have your files on the right places !! ;D
Long View
April 15th, 2008, 06:08 PM
{QUOTE-> Yes it not mentioned well but you should insist on keeping your hard drives cool it increases reliability. Thats why it good to have a front intake fan blowing over the drives keeps then cooler or wherever they be. I can fry an egg on top of my Raptors with no fan around but with a fan they feel cool to the touch. The casing act like one big heatsink. The long sides of the drives can be the hottest. A couple of these help http://us.st11.yimg.com/us.st.yimg.com/I/svcompucycle_1995_118980311 <-QUOTE}
How hot is hot and how cool is cool ? I am currently using a 6 year old segate that constantly reports 34c. Other drives seem to be ok at 40c + or even 50c +
laptps especially seem to run very hot
markymoo
April 16th, 2008, 09:56 AM
Long View (http://www.wilderssecurity.com/member.php?u=13562)
34C is considered safe and that is a common temperature. 55c has often be quoted as the maximum but that can't be true for all. I think there's abit of disinformation about it because nobody knows definite figures. It best to keep your hard drives cool as you can and if you can't keep them so cool at least keep them under 40C.
{QUOTE-> Manufacturers measure off quite a modest range of operating temperatures for hard drives, from +5 to +55°C as a rule, and occasionally to +60°C. This operating range is much lower than processors, video cards, or chipsets. Moreover, hard drive reliability depends heavily on their operating temperatures. According to our research, increasing HDD temperature by 5°C has the same effect on reliability as switching from 10% to 100% HDD workload. Each one-degree drop of HDD temperature is equivalent to a 10% increase of HDD service life. Once it rises by just 10°С – from 40°С to 50°С, the safety of its work becomes more than 2x less <-QUOTE}
Reading this thread may help.
http://www.silentpcreview.com/forums/viewtopic.php?t=7677
Huupi
April 16th, 2008, 11:36 AM
My take on it is,and is commonly known for many years that frequently turning on and off is far out the major cause for disk failure,its not the higher temperature as confirmed by the google teams findings.
One of their conclusions is that disks will fail,if they already have some inherent weakness,and it exposes itself mostly in the first weeks or months of use,beyond that time failures are fast diminishing.
I suppose that their servers are up 24/7 so temp. rise or fall is not an issue,but it surprised me that disks working on a higher temp.have no higher failure rate as compared with disk working on a moderate or cooler temperature.So get the message and kill the myth !!
It was a survey included thousends of disks (consumer disks !) over a long time so their finding are well substanciated.
Hairy Coo
April 16th, 2008, 06:45 PM
{QUOTE->
34C is considered safe and that is a common temperature. 55c has often be quoted as the maximum but that can't be true for all. I think there's abit of misinformation about it because nobody knows definite figures. It best to keep your hard drives cool as you can and if you can't keep them so cool at least keep them under 40C.
<-QUOTE}
Easier said than done!
Despite having my twin Seagates contained in their own cage with a 120mm fan blowing directly at them,its impossible to achieve an average of less than 40c,but then again Seagate say:
{QUOTE-> Seagate specifies in their spec doc 100129212b.pdf for the Barracuda IV -- "Ambient temperature: 0� to 60�C (op.), �40� to 70�C (nonop.)" <-QUOTE}
Brian K
April 16th, 2008, 08:01 PM
{QUOTE-> It was a survey included thousends of disks (consumer disks !) over a long time so their finding are well substanciated. <-QUOTE}
There were over 100,000 HDs in the sample. An impressive number.
In view of the data showing increased failure associated with lower temperatures, maybe we are doing harm by cooling our new HDs with fans. Another study will be needed to see if this is the case.
Long View
April 17th, 2008, 03:34 AM
{QUOTE-> Easier said than done!
Despite having my twin Seagates contained in their own cage with a 120mm fan blowing directly at them,its impossible to achieve an average of less than 40c,but then again Seagate say: <-QUOTE}
My Preference is for 2 of these http://www.quietpc.com/gb-en-gbp/products/harddrivesolutions/zm-2hc2
or http://www.quietpc.com/gb-en-gbp/products/harddrivesolutions/x-hdd-heatpipe
attached to the based of the case with a 120 mm fan in front. On one machine
2 Seagates run at 27c and 30c - no fan noise
Huupi
April 17th, 2008, 04:09 AM
its hard to revise your thoughts,i admit that after a long time living by the common mutual agreement that higher temp. will shorten the livespan,but the survey has not to be overlooked and i take these guys serious.
As always some feel comfortable living with the old myth,rather then opening their eyes for 'scientific' evidence to the contrary. ;) :)
edit) IF their findings are really trustworthy and the knowledge become gradually a part of the global awareness then LACIE would be glad and the cooling industry will take measures against it. trust me.
EASTER
April 17th, 2008, 04:15 AM
{QUOTE-> its hard to revise your thoughts,i admit that after a long time living by the common mutual agreement that higher temp. will shorten the livespan,but the survey has not to be overlooked and i take these guys serious.
As always some feel comfortable living with the old myth,rather then opening their eyes for 'scientific' evidence to the contrary. ;) :) <-QUOTE}
So whats the truth?
Is there a happy median temperature that preserves the lifespan of hard drives and can some heating actually have a bearing on extending their life or is this just a myth?
I run mine to extremes but i also keep 2 fans at full capacity and plenty of open area for proper air circulation.
EASTER
Eagle Creek
April 17th, 2008, 04:36 AM
Does anyone know how the position of the hard driver influences it's lifetime?
I've heard a story from someone who's computer had been horizontal for years. Then he moved the PC to another room and he put the computer vertical. Within a month, his had drive died.
Sounds like 'something' inside the hard drive got worn out and the position change was too much for it.
Huupi
April 17th, 2008, 05:01 AM
Ho Ho Easter these are not my findings,so calm down and read their rapport ones again,afterall you are talking about ONE drive,they are talking about over more then 100.000 drives ! IMO this make a difference.
To what degree the position of the drive has bearing on live expectancy,i don't know,to my knowledge they are made for both vert. and horizontal use.
markymoo
April 17th, 2008, 09:16 PM
{QUOTE-> Easier said than done!
Despite having my twin Seagates contained in their own cage with a 120mm fan blowing directly at them,its impossible to achieve an average of less than 40c,but then again Seagate say: <-QUOTE}
I am getting my WD at 37-41C with 120mm fan. It seems abit hopeful then. There are quite a few variables that come into play of hard drive failures. When i think back over 5-6 years i can't think of any hard drives failing. I still got my Raptors from 5 years ago working as good as ever. Thy haven't increased in noise either. I don't think it would of made a difference if they been running at 40C as opposed to 50C. They are good quality. The drives you buy today if you look after them generally last 3-5 years depending on usage. The common theme on my drive failures has been due to age or given it a knock, dropped somewhere down the line. If a drive can survive 3 years in good nick then it generally keeps going for longer. I know keeping motherboards cool prolongs there life. I apply the same principle keeping the hard drives cool. Run anything hot it gets damaged, run anything cold even subzero it doesn't and lives longer with less stress.
Meriadoc
April 17th, 2008, 09:58 PM
{QUOTE-> 2 Seagates run at 27c and 30c - no fan noise <-QUOTE}
Wow that's awesome Long View, that's defiantly my next purchase subject to finding out some more - (yeah easily satisfied I know ;D.)
{QUOTE-> ...how the position of the hard driver influences it's lifetime?..Within a month, his had drive died <-QUOTE}
I do have machines assembled with the harddisk fixed in horizontally and vertical in others so I'm interested to know anything on that also.
Hairy Coo
April 17th, 2008, 09:59 PM
{QUOTE-> Does anyone know how the position of the hard driver influences it's lifetime?
<-QUOTE}
Just read a review on a camcorder fitted with a vertically mounted 160gb hard drive,which was clamped to a motorbike handlebars for testing recording stability over rough conditions.
There were some sequence jumps,but once normal conditions returned-everything OK.
No damage.
Imagine the heat and stress under even normal camcorder conditions-but the drives just keep on working
A lot of cases like mine mount drives vertically.
In other words,you can mount them in any position you like,should have no effect.
The moving parts are hydraulically damped- etc-just dont drop or shock them enough to nullify the damping
Am convinced that drives have improved so much, that any normal usage whatsoever ,isnt going to worry them in the least.
Not worth worrying about-just use them:dry:
Hairy Coo
April 17th, 2008, 10:17 PM
{QUOTE-> I am getting my WD at 37-41C with 120mm fan. It seems abit hopeful then. There are quite a few variables that come into play of hard drive failures.
When i think back over 5-6 years i can't think of any hard drives failing. I still got my Raptors from 5 years ago working as good as ever. Thy haven't increased in noise either. I don't think it would of made a difference if they been running at 40C as opposed to 50C. They are good quality. The drives you buy today if you look after them generally last 3-5 years depending on usage. The common theme on my drive failures has been due to age or given it a knock, dropped somewhere down the line. If a drive can survive 3 years in good nick then it generally keeps going for longer. I know keeping motherboards cool prolongs there life. I apply the same principle keeping the hard drives cool. Run anything hot it gets damaged, run anything cold even subzero it doesn't and lives longer with less stress. <-QUOTE}
Well,I know sticking a damaged drive in the fridge is supposed to help(think its an old wives tale)-so sub zero may also help;D
Sounds like the danger period is the first 6 months,as manufacturing faults are discovered.
Generally if the drive survives that-a user should be OK.
Dont know if drives only last 5 years-your 5 year old Raptors are still working well.
But even five is enough for me.
In any case after this time period its generally worthwhile changing because of technology advances:thumb:
markymoo
April 18th, 2008, 12:11 PM
@Hairy Coo
That's right it doesn't matter which way you mount the hard drives can mount them upside down.
Sticking the hard drive in the freezer for a few hours or 24 really works! It contracts, shrinks all the parts. It works because anything misaligned can be corrected.
Yes 5 years is a good time to replace your drives. I know a few drives get noisy until untolerable to use.
lucas1985
April 18th, 2008, 12:42 PM
{QUOTE-> I do have machines assembled with the harddisk fixed in horizontally and vertical in others so I'm interested to know anything on that also. <-QUOTE}
HDDs only get damaged if you mount them inclined. Vertical mounted HDDs work fine.
Huupi
April 18th, 2008, 12:48 PM
OT though, but with the next generation of solid state drives,no mech. parts so this problem[mech. failure would be over],but for now they are small and expensive. i gues within a year or two they drop in price enough to make it interesting.
Eagle Creek
April 18th, 2008, 09:42 PM
Well, the SSD is gaining capacity very rapidly. It's too expensive to be interesting right now but I'm sure that problem will be solved within a couple of years.
vBulletin® Copyright ©2000-2009, Jelsoft Enterprises Ltd.
Copyright ©2002 - 2009, Wilders Security Forums