Best hard drive MBTF - Tilted Forum Project Discussion Community

kel · 11-28-2003, 07:54 PM

I did a search and didn't find a discussion on this particular question.

So the sounds from my linux box have finally come to bite me in the ass for ignoring them. 30 gigs of movies and TV I haven't yet burned to CD.

Now that I need to buy a new HD I have only one concern. What hard drives offers the greatest MBTF? Even better would be a hard drive that can warn me before it crashes.

I am not so concerned about speed or space because I don't use it as a desktop and the programs I develop aren't I/O or memory intensive.

sailor · 11-28-2003, 07:57 PM

Really, I dont think it matters too much, I think the only way you would see a difference is if you moved to SCSI or something that is meant for a hardcore mission critical environment. As far as I know, all of the manufacturers are about the same, and as MBTF is a pretty bad indicator of lifespan, its pretty much a matter of luck as to when your hard drive will die on you.

shakran · 11-28-2003, 08:36 PM

I've had a lot of different drives over the years. By FAR the most reliable has been Western Digital, to the point that I won't buy anything else now. I actually still have my first Western Digital drive, a giant 120 meg, which has been running every day for at least 2 hours (used to be much more when the 486 it's in was my main box) since 1993. That's twice the rule-of-thumb life expectancy for a drive. Sucker's even still quiet. The only reason I've EVER had to buy a new drive since I started using WD is 'cause I run out of room eventually. My quantum, seagate, maxtor, and IBM drives all died, but I have yet to have a WD crap out on me, and that includes all the WD's I've installed in clients computers as well.

Cynthetiq · 11-29-2003, 01:28 AM

M in the MTBF = MEAN which means there are numbers that fall in front of and after that number... it's not really an indicator of much...

Quote:

MTBF is the "mean time between failures," that is, the average elapsed time that passes before a failure occurs in a batch of drives under intense test conditions.

The initial inclination of those unfamiliar with the spec is to interpret it as the average expected lifespan of a single drive and either dismiss it as meaningless (bad) or assume this means their favorite drive will last decades (bad).

The SCSI-standard MTBF rating of 1.2 million hours, for example, does not mean that the average unit will run for 137 years before it fails. Likewise, the other extreme would be 1.2 million drives operating for one hour before one failed- equally unlikely.

Rather, MTBF is a product of a large quantity of drives (numbering in the hundreds or low thousands, perhaps) and the number of hours that such a batch runs before experiencing a failure. If a manufacturer places 1000 units to the test and on average manages to squeeze 50 days of operation out of the batch between each individual unit failure, that firm's product has achieved an MTBF of at least 1.2 million hours.

Sometimes, acceleration factors are used for calculating the MTBF of a drive. These are derived from standard statistical methods. Running the test at elevated temperatures beyond "normal," for example, will reduce the time needed to meet a certain number of test hours. Of course, the accuracy of this acceleration factor has a large effect on the final derived MTBF number.

That said, there's obviously some room for maneuver left in a "sweet spot" where firms may attempt to achieve target MTBF by either using more drives or more hours- there's no set unit count or elapsed time that we may regard as standard.

MTBF should be regarded as a minimum statement of reliability by the manufacturer. These days, no manufacturer will spec an enterprise-class drive below 1.2 million hours. Likewise, no firm will bother with MTBF less than 400,000 for a desktop-class disk. It is this consistency, rather than the spec's lack of meaning, that allows one to gloss over these claims.

http://www.storagereview.com/php/tik....php?page=MTBF

sailor · 11-29-2003, 07:21 AM

Yeah, basically, a company can easily manipulate MBTF to make their drive seem to last longer. Its not a good indication.

Shakran, thats pretty nice. At the risk of sounding like an old geezer though, I will say that I dont think they make them like they used to. It used to be that computers werent a real home appliance, and those that bought hard drives were willing to pay for good ones. Now, with everyone and their mother using computers and wanting moe space, quality has gone down. Remember recently when all of the manufacturers reduced their warrantees from three years to one?

Really the only company I would avoid is the IBM drives--they have been proven to have problems, and IBM admitted as much. Beyond that (WD, Seagate, Maxtor), they are all pretty much the same.

kel · 11-29-2003, 10:24 AM

I at least want to get a drive with fluid dynamic bearings. Do western digital drives use them?

sandeep · 11-29-2003, 03:25 PM

I'm not sure about the bearings issue, but I wanted to add that if it was critical data that you wanted to save, you could always mirror your drives on a RIAD setup. it would cost you a bit, but you could rest assured that the chances of both drives failing at onces is far from ordinary.

kel · 11-29-2003, 04:29 PM

It's not so much reliability as up-time. The last time I so much as physically looked at that linux box was a 4 months ago. Once I fix it up I don't want to look at again. I expect it to be there like death and taxes.

I move with relatively high frequency (I have a year and a half at my current location, before now I moved once or twice a year) and I suspect that helps explain why I have a history of frequent drive failures. They all last no more then 3 years.

I read that FDB motors handle shipping better.

I have been all over WD's site and they say that FDB is offered as an option on some drives, but when I try and find more details on it I come up empty. If you look at the specs of the drives it isn't mentioned. Maxtor however is very vocal about their use of FDB.

In a somewhat unrelated question, will older motherboards (PIII era) have difficulty recognizing large hard drives ie 160gig+?

charliex · 11-29-2003, 04:38 PM

SCSI drives tend to have better specifications lifetime wise, since they are deemed for professional rather than home use.

RAID 1 is good if you can afford to lose 50% of the total space, it'll often speed read performance, but will slow down write times, buts its a complete mirror so its a lot safer.

RAID 0 is only for speed and isn't a true RAID system, and won't offer any safety, in fact its worse than just having one drive.

RAID 2,3,4 arent all that useful, as they are inefficient , RAID 5 being the more popular for servers, but you need 3 drives minimum,a nd its complex to rebuild on a fault, unlike RAID 1.

beyond that the levels of RAIDs are really for servers and professional users since they require a lot of drives.

Two IDE's + a raid controller might be a better solution than a more expsensive SCSI drive with controller, depending on what you have now. I think RAID 1 is probably the best for someone looking for most data protection and home use at the cost an extra drive..

kel · 11-29-2003, 04:52 PM

What I mean is that buying two drives won't increase the life of the box in years. They will both fail within the same range most likely.

charliex · 11-29-2003, 05:42 PM

its possible, hard drives are curious beasts, so far (in recent years) i've only had western digital drives fail, i doubt though that two drives would fail exactly at the same time, unless they were abused physically, that way at least you could replace the bad ones, plus in a RAID setup the idea is to swap the drives out before they start to fail, so a RAID 1 with SMART drives, should give you lots of warning. so to keep costs lower, replace every other drive at 50% of the expected lifetime, say every 1.5/2 years or so.

SMART drives are supposed to be able to tell you when they are getting close to failing, you need to have the software and check it often to find out though. I did have a SMART drive that announced its own forthcoming demise, and it did indeed die later. you will probably need a hardware RAID 1 to hot swap drives though, i'm not aware of a software RAID that can hot swap.

not sure what you mean about reliability vs uptime, since one is dependant on the other ? do you mean you are ok with some partial data loss , rather than complete failure ?

Its the drive controllers + bios that control the 160GB, if the MB doesn't support it theres a number of solutions, some software absed, some hardware based, LBA48 support is required, you can always add a new controller card, some HD's even come with them once you go over 160GB, i believe the WD boxed do.

tritium · 11-29-2003, 11:05 PM

Quote:

Originally posted by kel
In a somewhat unrelated question, will older motherboards (PIII era) have difficulty recognizing large hard drives ie 160gig+?

Yes and no. Depends on your BIOS version. Flashed it lately? Could be a problem.

The second issue is your OS. Could be an old version of Windows ... might be a problem. Might not.

The IDE standards shifted to a higher density cable but I'm not sure where that fits in on the timeline of computer developments...

tritium · 11-29-2003, 11:06 PM

To add to the confusion ... I have a Western Digital Caviar 1.2GB drive die on me after 6 years of hard service. I like WD drives and absolutely would stay away from Maxtor. No opinion on anyone else. Maxtor is the KMart of the harddrive world, in my opinion.

alt0220bergeek · 12-01-2003, 10:57 PM

From my experience in industrial computing (think of your telephone company's equipment for running the damn phone system), only SCSI drives seem to still always be posting MTBF or RTBF. Consumer level drives (IDE and SATA) seem to be shifting to number of power cycles the drive can handle.

I don't remember what we have seen as far as which drive has the best numbers, but a good simple way around it is to run a RAID (either mirrored or go for the gusto and do a higher number, such as RAID 5, or 10 [yes it does exist, and it is WAY overdone]).

And even if you do back up the stuff from your hard drive to CD or DVD media (or tape), you might want to check your back ups every now and again (if it isn't easily replaced). Just a tip from my own misfortune.

11-28-2003, 07:54 PM	#1 (permalink)
kel WARNING: FLAMMABLE Location: Ask Acetylene	Best hard drive MBTF I did a search and didn't find a discussion on this particular question. So the sounds from my linux box have finally come to bite me in the ass for ignoring them. 30 gigs of movies and TV I haven't yet burned to CD. Now that I need to buy a new HD I have only one concern. What hard drives offers the greatest MBTF? Even better would be a hard drive that can warn me before it crashes. I am not so concerned about speed or space because I don't use it as a desktop and the programs I develop aren't I/O or memory intensive. __________________ "It better be funny"

11-28-2003, 07:57 PM	#2 (permalink)
sailor beauty in the breakdown Location: Chapel Hill, NC	Really, I dont think it matters too much, I think the only way you would see a difference is if you moved to SCSI or something that is meant for a hardcore mission critical environment. As far as I know, all of the manufacturers are about the same, and as MBTF is a pretty bad indicator of lifespan, its pretty much a matter of luck as to when your hard drive will die on you. __________________ "Good people do not need laws to tell them to act responsibly, while bad people will find a way around the laws." --Plato

11-29-2003, 07:21 AM	#5 (permalink)
sailor beauty in the breakdown Location: Chapel Hill, NC	Yeah, basically, a company can easily manipulate MBTF to make their drive seem to last longer. Its not a good indication. Shakran, thats pretty nice. At the risk of sounding like an old geezer though, I will say that I dont think they make them like they used to. It used to be that computers werent a real home appliance, and those that bought hard drives were willing to pay for good ones. Now, with everyone and their mother using computers and wanting moe space, quality has gone down. Remember recently when all of the manufacturers reduced their warrantees from three years to one? Really the only company I would avoid is the IBM drives--they have been proven to have problems, and IBM admitted as much. Beyond that (WD, Seagate, Maxtor), they are all pretty much the same. __________________ "Good people do not need laws to tell them to act responsibly, while bad people will find a way around the laws." --Plato

11-29-2003, 10:24 AM	#6 (permalink)
kel WARNING: FLAMMABLE Location: Ask Acetylene	I at least want to get a drive with fluid dynamic bearings. Do western digital drives use them? __________________ "It better be funny"

11-29-2003, 04:29 PM	#8 (permalink)
kel WARNING: FLAMMABLE Location: Ask Acetylene	It's not so much reliability as up-time. The last time I so much as physically looked at that linux box was a 4 months ago. Once I fix it up I don't want to look at again. I expect it to be there like death and taxes. I move with relatively high frequency (I have a year and a half at my current location, before now I moved once or twice a year) and I suspect that helps explain why I have a history of frequent drive failures. They all last no more then 3 years. I read that FDB motors handle shipping better. I have been all over WD's site and they say that FDB is offered as an option on some drives, but when I try and find more details on it I come up empty. If you look at the specs of the drives it isn't mentioned. Maxtor however is very vocal about their use of FDB. In a somewhat unrelated question, will older motherboards (PIII era) have difficulty recognizing large hard drives ie 160gig+? __________________ "It better be funny" Last edited by kel; 11-29-2003 at 04:38 PM..

11-28-2003, 08:36 PM	#3 (permalink)
shakran Tone.	I've had a lot of different drives over the years. By FAR the most reliable has been Western Digital, to the point that I won't buy anything else now. I actually still have my first Western Digital drive, a giant 120 meg, which has been running every day for at least 2 hours (used to be much more when the 486 it's in was my main box) since 1993. That's twice the rule-of-thumb life expectancy for a drive. Sucker's even still quiet. The only reason I've EVER had to buy a new drive since I started using WD is 'cause I run out of room eventually. My quantum, seagate, maxtor, and IBM drives all died, but I have yet to have a WD crap out on me, and that includes all the WD's I've installed in clients computers as well.

11-29-2003, 03:25 PM	#7 (permalink)
sandeep Insane	I'm not sure about the bearings issue, but I wanted to add that if it was critical data that you wanted to save, you could always mirror your drives on a RIAD setup. it would cost you a bit, but you could rest assured that the chances of both drives failing at onces is far from ordinary.

11-29-2003, 04:38 PM	#9 (permalink)
charliex Junkie Location: North Hollywood	SCSI drives tend to have better specifications lifetime wise, since they are deemed for professional rather than home use. RAID 1 is good if you can afford to lose 50% of the total space, it'll often speed read performance, but will slow down write times, buts its a complete mirror so its a lot safer. RAID 0 is only for speed and isn't a true RAID system, and won't offer any safety, in fact its worse than just having one drive. RAID 2,3,4 arent all that useful, as they are inefficient , RAID 5 being the more popular for servers, but you need 3 drives minimum,a nd its complex to rebuild on a fault, unlike RAID 1. beyond that the levels of RAIDs are really for servers and professional users since they require a lot of drives. Two IDE's + a raid controller might be a better solution than a more expsensive SCSI drive with controller, depending on what you have now. I think RAID 1 is probably the best for someone looking for most data protection and home use at the cost an extra drive..

11-29-2003, 04:52 PM	#10 (permalink)
kel WARNING: FLAMMABLE Location: Ask Acetylene	What I mean is that buying two drives won't increase the life of the box in years. They will both fail within the same range most likely. __________________ "It better be funny"

11-29-2003, 05:42 PM	#11 (permalink)
charliex Junkie Location: North Hollywood	its possible, hard drives are curious beasts, so far (in recent years) i've only had western digital drives fail, i doubt though that two drives would fail exactly at the same time, unless they were abused physically, that way at least you could replace the bad ones, plus in a RAID setup the idea is to swap the drives out before they start to fail, so a RAID 1 with SMART drives, should give you lots of warning. so to keep costs lower, replace every other drive at 50% of the expected lifetime, say every 1.5/2 years or so. SMART drives are supposed to be able to tell you when they are getting close to failing, you need to have the software and check it often to find out though. I did have a SMART drive that announced its own forthcoming demise, and it did indeed die later. you will probably need a hardware RAID 1 to hot swap drives though, i'm not aware of a software RAID that can hot swap. not sure what you mean about reliability vs uptime, since one is dependant on the other ? do you mean you are ok with some partial data loss , rather than complete failure ? Its the drive controllers + bios that control the 160GB, if the MB doesn't support it theres a number of solutions, some software absed, some hardware based, LBA48 support is required, you can always add a new controller card, some HD's even come with them once you go over 160GB, i believe the WD boxed do.

11-29-2003, 11:06 PM	#13 (permalink)
tritium Professor of Drinkology	To add to the confusion ... I have a Western Digital Caviar 1.2GB drive die on me after 6 years of hard service. I like WD drives and absolutely would stay away from Maxtor. No opinion on anyone else. Maxtor is the KMart of the harddrive world, in my opinion. __________________ Blah.

12-01-2003, 10:57 PM	#14 (permalink)
alt0220bergeek Psycho	From my experience in industrial computing (think of your telephone company's equipment for running the damn phone system), only SCSI drives seem to still always be posting MTBF or RTBF. Consumer level drives (IDE and SATA) seem to be shifting to number of power cycles the drive can handle. I don't remember what we have seen as far as which drive has the best numbers, but a good simple way around it is to run a RAID (either mirrored or go for the gusto and do a higher number, such as RAID 5, or 10 [yes it does exist, and it is WAY overdone]). And even if you do back up the stuff from your hard drive to CD or DVD media (or tape), you might want to check your back ups every now and again (if it isn't easily replaced). Just a tip from my own misfortune. __________________ Übergeek...