Seagate 1.5TB review
- Thread starter LiamC
- Start date
Mercutio
Fatwah on Western Digital
Given the number of 750GB Seagate drives I've had to replace, I don't think I'll be touching that one. Between my experiences with low-end 7200.10 80GB drives and "high end" 750GB drives, I am at the point where I'm ready to classify Seagate in the same category as Western Digital.
Gilbo
Storage is cool
Seagate disks have never done particularly well in the performance department. The 5 year warranty is nice though. And lately, they have been the ones pushing the capacities higher.
CougTek
Hairy Aussie
Seagate's ATA/SATA drives' reliability has been dreadful in my experience. I mostly sell Western Digital and their return rate is very acceptable. Maybe my suppliers handle W.D. drives more carefully than elsewhere around the Globe.
sechs
Storage? I am Storage!
Actually, I think it's Hitachi that's been doing that. They know how to fit that fifth platter, so they've been coming the market sooner with the higher capacities. This time around, however, it seems as if they've fallen too far behind in platter density to press that advantage.
Won't get me buy a Seagate, however.
Gilbo
Storage is cool
True, in the past it's been Hitachi. Seagate has only started pulling ahead lately.
Fushigi
Storage Is My Life
*shrug* Mine is working fine so far.
Mercutio
Fatwah on Western Digital
The only problem is that one drive isn't a very large sample.
In the grand scheme of things the 30 or so 750GB Seagate drives that have passed through my hands certainly isn't either, but between my poor experiences with ~10 of those 30 drives, and with 31 of the 220 80GB drives I worked on with my summer computer building classes, I feel like I've got an unpleasant picture of Seagate desktop products. Mostly I'm talking about -AS rather than -ES products, but I've sent back -ES drives too.
Basically, I'm thinking that Seagate has run out of good will, and that 1.5TB is an awfully large chunk of data to trust with them.
In the grand scheme of things the 30 or so 750GB Seagate drives that have passed through my hands certainly isn't either, but between my poor experiences with ~10 of those 30 drives, and with 31 of the 220 80GB drives I worked on with my summer computer building classes, I feel like I've got an unpleasant picture of Seagate desktop products. Mostly I'm talking about -AS rather than -ES products, but I've sent back -ES drives too.
Basically, I'm thinking that Seagate has run out of good will, and that 1.5TB is an awfully large chunk of data to trust with them.
Fushigi
Storage Is My Life
I admit the sample size is too low to be meaningful. I don't go through enough drives personally to ever have a meaningful sample size. Professionally, I only deal with 15K SCSI disks, moving to SAS at some point, and while those are pretty much all Seagates their falure rate has been something like 1 every 9 months across 96 drives. They've proven a lot more reliable than the 10K drives from older servers.
With all the doom & gloom pronouncements that not all of the drives die right away. Of course this drive hasn't been in service that long, but the Seagate 750 in a different machine has been fine for something closer to a year now.
The 1.5TB model is perfect for my Home Server. The main reason for the capacity is to store backups of the workstations. Other data we store there specifically - music, photos, etc. - is duplicated to a Samsung 500GB unit via the WHS Folder Duplication capability.
For my personal PC farm, to be honest, for the past 5 years or so all drives have been fine. Only one old 20GB PATA Samsung failed on me. The various WD, Maxtor, etc. drives were all fine when decommissioned & replaced with SATA units. I've only been using SATA since May '07 but haven't had any issues with any drives so far. That includes the afore mentioned Seagates, 3 500GB Samsungs, and an 80GB Fujutsu notebook drive.
Again over the past few years for my work notebooks, the Dell-supplied Hitachi seems to die around 18 months while the rpelacement lasts indefinitely. That's happened twice so far; we'll see what happens next summer when my current notebook hits that age.
With all the doom & gloom pronouncements that not all of the drives die right away. Of course this drive hasn't been in service that long, but the Seagate 750 in a different machine has been fine for something closer to a year now.
The 1.5TB model is perfect for my Home Server. The main reason for the capacity is to store backups of the workstations. Other data we store there specifically - music, photos, etc. - is duplicated to a Samsung 500GB unit via the WHS Folder Duplication capability.
For my personal PC farm, to be honest, for the past 5 years or so all drives have been fine. Only one old 20GB PATA Samsung failed on me. The various WD, Maxtor, etc. drives were all fine when decommissioned & replaced with SATA units. I've only been using SATA since May '07 but haven't had any issues with any drives so far. That includes the afore mentioned Seagates, 3 500GB Samsungs, and an 80GB Fujutsu notebook drive.
Again over the past few years for my work notebooks, the Dell-supplied Hitachi seems to die around 18 months while the rpelacement lasts indefinitely. That's happened twice so far; we'll see what happens next summer when my current notebook hits that age.
I really have had no problems with the Seagate AS drives. They are not the fastest, but they are good enough and much better than the WD's that run at 5400 which is their price competitor. However, I'm very displeased with the 1.5's slow write times.
I'm starting to wonder if the perpendicular drive technology that is driving these giant sizes is having performance issues that has not been discovered and/or discussed sufficiently by the media.
I'm starting to wonder if the perpendicular drive technology that is driving these giant sizes is having performance issues that has not been discovered and/or discussed sufficiently by the media.
Mercutio
Fatwah on Western Digital
I had another 750GB -AS die this morning. It was actually kind of spectacular, and judging from the burns on the drive cage there was either fire or arcing. I smelled it, and there's visible damage to the smaller of the two ICs and some of the traces.
I checked my other 750GB -ASes and that chip is EXTREMELY warm on all of them. I'm pulling all those drives out of service now.
I checked my other 750GB -ASes and that chip is EXTREMELY warm on all of them. I'm pulling all those drives out of service now.
What 1.5TB Western Digital drives? Are there any such drives besides the Seagate?
ddrueding
Fixture
I'm still running an array of 12 750GB Seagate drives, but I have my first two RMAs sitting on my desk (one kept dropping from the array, the other with a broken SATA connector).
I think I'll be moving back to Samsung myself.
I think I'll be moving back to Samsung myself.
More possible bad news: Problems with Seagate 1.5TB drive and Linux or Raid.
udaman
Wannabe Storage Freak
- Joined
- Sep 20, 2006
- Messages
- 1,209
Heat is the enemy of most electronic devices
Dells are notorious for cooking laptop drives, they just are not cooled adequately, hence many failed drives. Not sure why a replacement would last indefinitely, only you may be putting in a newer drive that has a slightly lower dissipation owing to greater efficiency, lower power consumption?
Could be, but I have another theory. It's a general manufacturing theory that applies to electronics, computers and extends in a like analogy to pretty much all products. You get what you pay for...and manufacturers facing extreme competition (much more so in todays fierce global economies) will use the lowest cost parts, that 'appear' to meet design spec's. Though I believe those parts are never adequately tested before pressure/demands of fast to market needs, limit amount of testing that can be done.
Cheap parts are the achillies heal of all manner of products.
My parents are elderly, in such weak/poor health they can no longer do many of the things that require even modest/minimal physical activity. Instead of them paying $50/hr or more for a plumber to do a simple replacement of a garden hose, 1/2in & 3/4in brass faucet valve. I do it for them. I'm shocked at the abysmal quality of rubber washers in these water valves. I recall replacing a washer on a made in the USA Champion brand brass valve...once the standard bearer of quality brass valves, water fittings. Last time I bought one of these Champion valves, it was a anti-siphon type, cost more for that feature...about $15 say, 5+yrs ago. But it started leaking, would drip, drip, drip after just a few years use. I tried to take it apart to replace the washer, which would have cost say 1/2 a dollar. Couldn't do it with the tools I had.
Bought an another American made brass valve at a different hardware store since the store I bought the Champion at, had switched to a less expensive made in China brand, with poorer quality handle on the shut-off. Other American brand that, while not anti-siphon, still costs $10+ at teh time a few years ago, best I can remember, the plastic barcode scan tag still attached around the stem of the valve. It started to leak also, damn!
Went to the store which still carries that line of valves, asked the salesman if he could give me the proper size replacement washer, as the new valve now costs $14. WTF, no reason to replace a relatively new brass valve because the crappy washer they are now using in these valves, only lasts for a few years before they start to leak. They guy took a new valve assembly back into the tool room, put it on a vice and tried to use wrenches to take it apart to figure out the size of the washer. He gave up after completely marring the brass surface with wrenches, unable to get it to come apart. You can with lots of labor, I have found unscrew the securing screw in the center of the washer, then with must fuss, and screw drivers, fold/bend the washer in half, remove it and replace in same manner...total PITA.
I told the guy, forget it, I'll just have to go buy the $6.99 cheaper Chinese valve, and replace that every few years when it starts to leak. massive waste of materials/manufacturing, all because every valve, it seems, is using inferior rubber washers. WTF, it would cost an additional dollar to and already expensive brass valve, to get a high quality silicon rubber washer...which would last for more than a few friggin years!!! Think of people who don't want to put up with a leaking garden faucet, lack skills or motivation to get required wrenches, and end up paying for the plumber to do this job...every few friggin years, all because of the use of inferior inexpensive parts. Pisses me off.
/end rant
The Google white paper on hard drive failures in their data centres from a year or two ago did have a very, very good sample size, and they came to the conclusion that heat was not an issue in drive failure...
First I've heard that one. Do you have any references?
...rant
Agreed. This replace rather than repair mentality is just wrong. Certainly so if we are (and should be) trying to reduce resource usage.
Stereodude
Not really a
I don't buy it. I have seen the curves and formulas used to determine lifetime of electronics. The temperature is a key value in the formula to determine lifetime.
My observation is that electronics failure is not the typical failure mode of HD's. It does happen sometimes and that's why people recommend replacing the circuit board as a solution to some problems. My personal history has been that that has never worked for me, but I've only tried it a few times.
Read the paper. It's hard to argue with (the) data
Some relevant quotes:
..."The data in this study are collected from a large num-
ber of disk drives, deployed in several types of systems
across all of Google’s services. More than one hundred
thousand disk drives were used for all the results pre-
sented here. The disks are a combination of serial and
parallel ATA consumer-grade hard disk drives, ranging
in speed from 5400 to 7200 rpm, and in size from 80 to
400 GB. All units in this study were put into production
in or after 2001. "...
100K consumer drives. A large enough sample.
..."Our key findings are:
• Contrary to previously reported results, we found
very little correlation between failure rates and ei-
ther elevated temperature or activity levels.
• Some SMART parameters (scan errors, realloca-
tion counts, offline reallocation counts, and proba-
tional counts) have a large impact on failure proba-
bility."...
and
..."One of our key findings has been the lack of a con-
sistent pattern of higher failure rates for higher temper-
ature drives or for those drives at higher utilization lev-
els. 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 reliabil-
ity in the context of a professionally managed data center
deployment."...
http://research.google.com/archive/disk_failures.pdf
..."The data in this study are collected from a large num-
ber of disk drives, deployed in several types of systems
across all of Google’s services. More than one hundred
thousand disk drives were used for all the results pre-
sented here. The disks are a combination of serial and
parallel ATA consumer-grade hard disk drives, ranging
in speed from 5400 to 7200 rpm, and in size from 80 to
400 GB. All units in this study were put into production
in or after 2001. "...
100K consumer drives. A large enough sample.
..."Our key findings are:
• Contrary to previously reported results, we found
very little correlation between failure rates and ei-
ther elevated temperature or activity levels.
• Some SMART parameters (scan errors, realloca-
tion counts, offline reallocation counts, and proba-
tional counts) have a large impact on failure proba-
bility."...
and
..."One of our key findings has been the lack of a con-
sistent pattern of higher failure rates for higher temper-
ature drives or for those drives at higher utilization lev-
els. 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 reliabil-
ity in the context of a professionally managed data center
deployment."...
http://research.google.com/archive/disk_failures.pdf
And just stumbled across a 750GB drive roundup
http://www.xbitlabs.com/articles/storage/display/750gb-hdd-roundup.html
Hitachi made and impression--Long live the Deathstars
So did Samsung, and selected Seagate & WD models
http://www.xbitlabs.com/articles/storage/display/750gb-hdd-roundup.html
Hitachi made and impression--Long live the Deathstars
So did Samsung, and selected Seagate & WD models
Stereodude
Not really a
I've read the paper, and I don't think it's hard to argue with the data. First for drives in service for 4 years or longer the hotter drives do have a higher failure rate. Second, I'm not sure I'd consider the "hot" temperatures they saw hot. It looks like only about 10% of their drives (total) were over 40C. If I recall correctly, a 7200 RPM drive in a poorly ventilated case will get warmer than the temperatures they saw.
Mercutio
Fatwah on Western Digital
I can tell you from experience that drives that are more than about three years old that have been used in a server are more likely to fail, no matter what you do.
I make it about 17% 40 or over. Define hot. This is consumer grade hardware we are talking about, and a quick scan of WD spec sheets reveal operting temps (consumer) of 60 or 65 max. Sure operate a drive in summer flat out in a porrly ventilated case and the drive will be operating at the top end or above it's rated environment. Then all bets are off. You can't complain of failure in that case.
And what the report stated was that their was a mildly inflated failure rate with old, hot and hard working drives, but other factors had a greater effect/higher incidence of failure. This isn't a black/white situation.