IBM 120GXP Revisited 360
Andrew sent us a link to an article about the
IBM 120gxp controversy.
This is about the fact that the drive has been declared unfit for server use,
and to back that up, IBM says you should only use it for 333 hours a month.
This is a good summary of the issues and worth a read.
8 hours a day? Please... (Score:4, Informative)
Furthermore, the DeskStar isn't intended to be a server part - IBM makes the UltraStar for that.
So, in essence, it's buyer-beware with OEM parts. Just like with the ATI video card debacle - You're buying parts that aren't intended for *you* to use. It's your fault if you're tryin' to skimp a couple of bucks out of IBM/ATI/whoever by buying on the grey market.
Now, that said...it's pretty fscking ridiculous to be making these drives and all but marketing them as the fastest ATA drives on the planet. That's practically hyping it up to the enthusiast market right there. And I really think it's asinine to expect these drives to *only* be run 8 hours a day. Factoring in the average lunch break when the computer will most likely get left on, that means that the drives are generally running out of spec on a regular business day in your average workstation.
Re:Yes, yes, we get the point. (Score:1, Informative)
Why I will never buy any GXP drive (Score:2, Informative)
Scary.
Personal Experience (Score:2, Informative)
Hopefully, the replacement they sent will last more than 6 months. But, just in case, I have a Maxtor 60GB in place as a backup. At least this time, if it goes down, I won't have to wait for the replacement.
At first they are quiet (Score:2, Informative)
Re:8 hours a day? Please... (Score:3, Informative)
Seems to be /.'ing (article text) (Score:1, Informative)
The IBM 120GXP Revisited
The revelation that IBM's 120GXP line of drives was only rated for 333 hours of usage per month kicked off a storm of controversy and discussion both here at VH and at other websites as well. With little actual data to go on, however, speculation has run rampant and official response has been scarce. VH launched its own investigation into the issue with the goal of examining IBM's claims regarding the 120GXP's optimum usage, whether such claims were sufficiently documented, and how consumers may wish to respond to the situation. We've organized this article in a question-and-answer format to allow for simpler organization and easier reading, with the questions themselves designed to walk a reader through the situation.
IBM, unfortunately, would not provide additional information or official comment on the 120GXP situation, despite repeated phone calls and requests for data.
Is there a reliability problem with IBM hard drives?
This isn't an easy question to answer, especially with IBM withholding comment. While large numbers of readers responded to the questions I posed regarding drive reliability, their emails present very different pictures. Some of you swear by IBM drives and their reliability, while listing many of the Seagate, Maxtor, or WD drives you've seen fail in both a corporate and a consumer setting, while other readers had horror stories of seeing IBM drive after IBM drive bite the dust.
Based on the emails and feedback we received from earlier stories, however, the IBM drives seem to have one distinguishing characteristic--they fail faster. Most of the emails and commentary we received indicated drives that survived only a few months past their purchase date. My own personal experience mirrors this--of the three 75GXP drives I've replaced for customers, two of them were less than a year old, with the third being just shy of eighteen months.
If you take into account the fact that IBM is currently facing a lawsuit alleging that the 75GXP line is defective, it seems fairly clear that, at the least, IBM's 75GXP line is suffering from problems. How widespread the problems are, which drive models they affect, and what IBM is doing to fix them, however, is not clear. It's been suggested that all of the defective drives have shipped from a certain factory, but Big Blue has done nothing to confirm or deny this, or even acknowledged that a problem exists.
How does the 333 hour limit fit into the current situation, and what does the limit mean?
Last week, IBM dropped a bombshell on the hard drive community when Storage Review published a conversation between IBM and "a long time SR participant." The conversation stated that these drives were unsuited for any type of serious server role and should not be used as such.
To some, this is simply IBM covering the bases of liability by stating the drives should only be used eight hours a day, but means nothing else. Others of you have expressed grave concern that the newly-emphasized 333 hour limitation is, in fact, a tacit admission by IBM of problems with the 120, 75, and 60 GXP drives.
While there are arguments to be made for both sides, the bulk of the evidence points to the latter rather than the former. Several websites (including Tweaktown) have published articles indicating that the 75 (and possibly even the 60 GXP) drives are failing because of excess heat. Certainly it's true that using the drive less would be one way of keeping it cooler over its total life, allowing for greater reliability.
Also of note is that IBM has never emphasized this 333 hour-per-month usage rate before on any of its products. While the specification exists in the technical literature for the 120, 75, and 60 GXP drives, websites reviewing these drives for months have recommended them for placement in low-end servers. IBM's own press releases have targeted the drives for these areas as well. It's extremely odd that a company would both encourage websites to review its drives in a low-end server environment, target them into that market, and then suddenly pull an about-face and claim the drives should not be used in such arenas.
One argument used to support the idea that the 333 hours-per-month is merely a liability trick is that companies like VIA ship their C3 processors with a heatsink and fan, yet publicly demonstrate the chip running with only a heatsink and are known to encourage the use of the processor in such a configuration.
The problem with this argument, however, is that VIA doesn't state that the chip can only be used for a fraction of its original time OR at a fraction of its original speed in exchange for removing the fan. By stating that the drive should only be powered for 333 hours per month, IBM is limiting usage to about eleven hours per day. In other words, VIA sells the chip with a heatsink and fan, but also deploys it in a heatsink-only configuration with no loss of performance. The fan may be there, but only for liability reasons.
By stating that their drives should only be used for eleven hours per day, IBM is drastically limiting the performance of the drive (as compared to its competitors) as well as its appearance in the market. The ability of the VIA chips to run with less cooling than expected is a tremendous positive--but for IBM drives to be perceived as only recommended for eleven hours a day when their competitors make no such recommendations about their own product is a tremendous negative. The argument that the two situations are comparable, therefore, is false. One company is demonstrating a product that exceeds market expectations, while the other is demonstrating a product that fails to meet them.
Part of the confusion regarding this situation could be resolved if IBM would clarify whether the 333 hours-per-month stipulation referred to hours the drive was powered on, or hours the drive was engaged in reading/writing. Once again, however, the company has not deigned to comment on the situation.
Has IBM properly disclosed the operating conditions of the GXP line of drives?
The one comment I DID get from IBM when I brought up the 333 hours-per-month operating time restriction on the 120GXP is that this stipulation has, in fact, been present on ALL of the GXP lines. We investigated their claim and found the following:
*
The restriction is not noted on any of the GXP drives itself.
*
No documentation was included that mentioned this monthly restriction to purchasers of the GXP line at either the consumer or business level.
*
No information mentioning the GXP's recommended hourly rate is discussed in any press release related to any GXP product. Furthermore, the drives are expressly and explicitly targeted to high-end enterprise buyers.
*
No data about the restriction is present on IBM's Deskstar homepage.
*
There IS, in fact, information about the various restrictions present on the specific pages for the drives themselves, but the information is in different places for each drive.
*
For the 120GXP, the restriction is noted in the two-page "Data Sheet and Specification" document under the "Reliability" section.
*
Neither the 75GXP nor the 60GXP have the 333 hour-per-month specification mentioned in their own versions of that document, however. The 60GXP lists this setting only in its "Functional Specifications" document--a hefty 195 page engineering-level PDF. The specification in question is located 'prominently' on page 50 in a relatively small section. The 75GXP, on the other hand, does not have a "Functional Specification" link and does not mention the limitation on its data sheet either. I was unable, in fact, to even FIND mention of such a limitation for this particular model.
So, we're left with the following situation: A limitation IBM claims is specified on all three drives isn't mentioned in the shipping documentation, on the drive itself, or in any press release. The specification, in fact, is mentioned only in three different places on three different websites, and is only easily found on one. Furthermore, the limitation in question is placed on drives IBM markets as being "enterprise solutions", suitable for high-end workstation use for multimedia and graphics presentations. Its not as if these are IBM's budget line of drives, after all--these drives are (theoretically) the top of the line models.
This does not strike me as proper disclosure of a drive limitation. Even if the drive limitation stretches beyond the IBM Deskstar line and into its competitor's products as well, the company has essentially switched performance metrics without informing anyone of the change.
The best example of a proper metric-switch is AMD's recent decision to use model numbers instead of MHz. When they made the change, AMD highly publicized it and openly disclosed their measurement data, their specifications, and the reasons for the metric switch. Whether people agreed or disagreed with it, the quantifiable data was laid on the table for examination.
In this case, even if IBM were to argue that other IDE hard drives from other manufacturers suffer from the same limitations; their failure to explain and discuss this new "hours per month" metric has drastically hurt their drive's perception in the market.
This situation actually reminds me from a scene in the popular book "Hitchhikers Guide to the Galaxy" where Arthur Dent (who's house is about to be destroyed) questions a public official about when the plans to destroy it were made and how he was informed. The following quote illustrates the situation quite well:
"But Mr. Dent, the plans have been available [and on display] in the local planning office for the last nine months."
"On display? I eventually had to go down to the cellar to find them."
"That's the display department."
"With a flashlight."
"Ah, well, the lights had probably gone."
"So had the stairs."
"But look, you found the notice, didn't you?"
"Yes," said Arthur, "yes I did. It was on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying "Beware of the Leopard."
Sound familiar?
Re:8 hours a day? Please... (Score:5, Informative)
Oh please, probably 80% of the drives sold online (where most enthusiasts get their drives) are OEM. The average person buys a retail Maxtor at CompUSA, whereas most people who have some idea of what it is they're buying (aka enthusiasts) get the drive without all the extra cardboard/paperwork from either a local computer shop or from the internet. The drives purchased seperately as OEM are the workhorses. All my drives (6 of them) are OEM and they spin full speed 24/7 with maybe an hour of maintenance downtime every 3 months.
Repair Depot Comments (Score:5, Informative)
"SCSI" was his response. "Oh sure," he said "there are fewer of them out there -- but we hardly see any at all."
One of the other interesting things he told me is that the drives included in Quantum's Snap Server appliances, despite being IDE, are *NOT* drives you can buy off the shelf. And he hasn't gotten in a single Snap Server drive in the two years they had been selling them.
I switched from Quantum to IBM drives at the time (ugh!) but had the forsight to put them all in a RAID-1 configuration. We've sent three DeathStars back for repair so far. The good news? IBM had replacements to us in under a week.
I recently switched from IBMs to Maxtor (making sure I wasn't buying Quantum's old stock) and have already had one of their 80Gb drives fail. For the record they are not as responsive as IBM in the RMA department.
So what's the answer folks? You get what you pay for. If you care about your data buy an Adaptec 1200A RAID-1 controller and two drives, or spend the money on a SCSI controller and SCSI drives. So far I haven't found any IDE drive vendors that can sell you a reliable drive (I have dead fujitsu drives around here as well, but must admit that I still haven't tried Western Digital.)
Re:So what are they good for? (Score:3, Informative)
Re:Even 333hr per month is pushing it (Score:4, Informative)
Same thing with Maxtor.
I had a 40GB Maxtor (D540X) die on me three months after buying my computer. They sent me an advance replacement, and two months later that one died. When I called to RMA *THAT* one, and to complain about the short lifespan, they asked if I wanted a "new build instead of a refurbished drive" this time. My response was... HELL YES!
Given that comment from their customer service rep, it sounds like most drives under warranty are replaced with refurbs.
Same 333 Hours in 60GXP and 120GXP (Score:2, Informative)
60GXP Spec Sheet, Pg 50 of 209 [ibm.com]
120GXP Spec Sheet, Pg 2 of 2 [ibm.com]
There's probably one of these for the 75GXP line. But I think it is implied from the get go already, if the clicking doesn't kill you first.
What ever you do, just don't put two of these babies in a TiVo! [tivocommunity.com]
Re:8 hours a day? Please... (Score:2, Informative)
Cutriss wrote:
Actually if you read the PDF datasheet from IBM [ibm.com] (linked to from here the Deskstap 6GXP page [ibm.com]) it does have the figure of 333 hours use per moth for 5 years on page 50.
I'm rather pissed off at this since I brought one of these drives a few weeks ago and it's running my desktop machine, which is on 25/7... it's OK so far but it would be a pain if it dies, but since /home/ is NFS mounted it wouldn't be the end of the world.
If I had read about this before I went out to get a new drive I would have brought a brand other than IBM.
It's not only 75GXP & 120 GXP (Score:2, Informative)
Of course this doesn't compare to horror stories of people where all 4 IBM disks in a RAID dies within a month. I think the whole GXP line smells bad.
D.
Re:Repair Depot Comments (Score:1, Informative)
And cheap SCSI drives do not cut it. The reason SCSI drives are more solid is because you're paying more for nicer drives. You can't run out and get the cheapest SCSI drive on Pricewatch -- you need to pay for a higher end model.
7200 rpm isn't worth it. (Score:1, Informative)
Drive reliability ratings (Score:5, Informative)
Here's a resource I've been watching lately. If anyone has similar things (published reports of reliability from places that deal with dead drives) please follow up to this message.
http://www.driveservice.com/bestwrst.htm
Re:Reliability has a cost (Score:2, Informative)
A poster above has had many UltraStars tank; the lesson being that all brands and types of hard drives can fail. Don't think you're immune. If you haven't been bit yet, chances are good that eventually you will be.
Deskstars also have a three year warranty. You could check before alluding a falsehood.
Smugness is not a substitute for insightful commentary, and is even less appreciated when a lack of logic and facts are present.
I hate to repeat myself but... (Score:2, Informative)
Well lets see if we take your math further that;
333hours/mo * 12months *5 years (off the graphic you posted) = 19,980 hours total. That's Horrible!
([H]ardOCP http://www.hardocp.com had posted a graphic from IBM's documentation that said the expected life of the drive was 5 years, hence the 5 above)
Lets see WesternDigital rates their 120G at 500,000 hours on the bottom of ; http://www.wdc.com/products/current/drives.asp?Mo
Lets compare a High end, high quality drive the Cheetah X15 it has a MTBF of 1,200,000 hours!
Ok lets play fair and compare it to something a little older like it is how about a Seagate ST4766E 667meg hd circa 3/29/90 (according to the bad sector sticker on the drive). According to Seagates web site ( http://www.seagate.com/support/disc/specs/esdi/st
Lets try something older. How about an old ST 225, nope that's got a MTBF of 100,000 hours. http://www.seagate.com/support/disc/specs/mfm/st2
Hmmm do you think if we grind enough IBM drives up we can make some lemonade?
They are reliable! (Score:2, Informative)
I have on 60GXP from Singapore(?) and & 75GXP. I spaced them properly, and placed holes where they are located, so that powersupply fan would pull air around them. I never had any problems yet!
IBM excellent drives, if they be more understanding to the issue, it would be great, but making people hate great product, instead of instructing them how to work around the problem, that most other drives have is to say the least is counter productive. But then thats what reading slashdot is all about anyway