Slashdot Mirror
Enterprise-class ATA Drives
dfung writes "This has been mindlessly discussed many times before here, but Western Digital has now introduced real enterprise-class ATA drives with SCSI-like performance specs and 30% lower price. So now you can buy a real 10K rpm ATA drive. Interestingly enough, they mention the reason for the traditional difference in price between ATA and SCSI which I never have seen mentioned here - it has to do with testing costs, not controller electronics|platter quality|etc. Another interesting tidbit is that 160 million ATA drives were sold last year. I saw about 2 million of them stacked up in the aisles at Fry's Electronics yesterday, but that sure is a lot of drives."
Been waiting for this for ages =)
- It's going to be a real plus when running IDE-raid solutions too. Esp. if you compare the prizes to the SCSI solutions.
GO WD go, hopefully, these drives will be of higer quality than their recent IDE drives that have been breaking by the ton a week too...
-- 040
The Xserve and the fiberchannel Xserve raid would go nice with these new drives...
||| I still can't believe Parkay's not butter.
Since the price difference is only 30%, SCSI should be the obvious choice for server type tasks... considering all the other benefits of SCSI. IDE seems kinda hackish in comparison.
Save your wrists today - switch to Dvorak
Now if only they make a 200G version with the 8M cache, gotta love those special edition drives.
Horror & SciFi Erotic Nudes
1. Is there anything to relate drive geometry and the interface?
2. Testing time is a function of prodn. capacity. Obviously there'd be 10 times as many ATA drives as SCSI.
3. Spindle speed and drive interface - any connection?
More marketing spin here than drive spin. Probably enough to win the desktop PC market. If MS can spin, WD can do better. What next? ATA-XP drives specially tuned for XP??
God is an Anonymous Coward....jkrise
If you keep throwing chairs, one day you'll break windows....
If this had the same capacity as the "desktop" IDE drives, say 120+ GB then we would be talking. We don't use any drives SCSI or otherwise below 60 GB for our servers.
Remember, there are no stupid questions. But there are a lot of inquisitive idiots.
hmm; don't think of it that way. it's simply not possible for people to go and plug in stuff by hand on a cable and test them. usually. that's wayyyyyyy in the end anyway - intergration testing in software people's terms?
;). after each "scan burst" you toggle the clock, and sequentially read out all the registers to see if the chip did what you wanted it to do.
I would *suspect* that scsi chipsets have more things to be tested than ATA ones, since as you may notice, they are supposed to work with 15 devices OR by themselves, possibly providing onboard termination or not.
testing often starts at the wafer stage (where each chip is probed and marked, failed ones are crushed, oftenly), and again when chips are packaged - usually speed sorted / repaird (if possible - a lot of memory devices support repair) at this time. After that, integration testing is actually EASIER because this is when you have a whole set of firmware commands to work with, etc.
Frequently chips have dedicated testing commands, though (that you don't get to know), so things are not completely dire. most flash memory have test modes, for example, where if you put in a code sequence it will write the entire array into, say, a checker board pattern. This is to avoid massive delays of half microsecond writing each location, sequentially. Logic chips (like, say, scsi chipsets) usually have a different challenge - they have embedded subsections, often cache, that you don't have access to directly.
now, to get "into the chip" you will have to sequentially put in the test patterns / vectors into special registers that reside on the lines that run between each embedded component. one register at a time (usually sequentially through a few (dozen or less) pins. testing is expensive, but pins more so
this gets back to the scsi being harder to test - probably the control chipsets are more complex. I can't imagine the mechanical sections being any different (besides the 15krpm ones, anyhow) - generally when something have to communicate with a bunch of other things (like scsi) versus just a few (ATA), the former is more pain in the butt testing wise.
oh, btw - more PIN is also another factor to costs. testers have a limited number of pins, so if you have more pins to test, you test less per turn. can't speak authoratatively on the pincount of drive controller chipsets... just FYI here.
side note: one thing you realize after being in testing is that semiconductor manufacturs often (or, sometimes - depending on the manufacture) puts a LOT of margin into their chips. when they say the chip is rated 75 degrees C, they really mean 75 degrees because the chip was TESTED at that temperature.
ok. long rant... gotta stop now.
My life in the land of the rising sun.
It is always interesting to see all the jumping through hoops to defend the use of SCSI. But I think it is all bullshit.
Every manufacturer could, at any time, start producing a diskdrive that has the mechanical and head/servo electronics of an existing SCSI drive integrated with an ATA bus interface. It would have the reliability of the SCSI drive, and assuming that manufacturer has experience in ATA electronics there is no reason to assume that it would have problems on that end.
No need to have it in the market for 5 years to prove reliability. Disk drives are not even in the market for such a long time.
No, they just want to sepatate two different price categories and don't want to blur that gap by offering drives with features from both sides.
Heh. Vastly increased heat production. 10k RPM drives get quite hot!
http://rocknerd.co.uk
Dont forget more noise. This usually is not an issue in a server room (there's so much noise already, who would notice), but it is if you want to use these drives in a workstation.
You can't go by other's opinions when it comes to computer products anymore. People get burned once by one manufacturer and they will lambaste it all over the place. Same goes for the question "where do I buy product X from?". Look around and you'll find a horror story about EVERY manufacturer and EVERY place of business.
Just ignore other people, and go by price/specs/warranty.
When you say 'serious server', you have to qualify it. For systems where everything's going to be loaded into memory, and only logging from ONE application, you're fine with an X1 or V100. [eg, NTP, DNS, DHCP].
Now, for anything that requires file access (HTTP, LDAP, NNTP, SMTP, or even just running multiple apps on a single box), you're going to want to go to SCSI for the benefits of tagged queuing (yeah, I just posted this link on a seperate thread, so the link's redundant, but the message it's supporting is different)
As you said, ATA is fine for desktops, as for the most part, the person's oly doing one thing at a time, however, if there's major disk I/O (video/audio editing), you start getting to 'workstation' class, and could get a performance increase out of SCSI or FC-AL.
As with any engineering or tuning process, you need to know what the characteristics of the system are before you can make a decision. If the process is bottlenecked by CPU, memory, or network I/O, the disks may not have an impact -- however, upgrading one of the other items may suddenly create a need for a better storage architecture.
Build it, and they will come^Hplain.
36GB? Give me a call when the 160, 200, or 320's are in that range for price and reliability. Its just more cash scamming. I've seen 36GB drives for as low as $50...the testing makes the price go up stuff is pure corporate %&***^)(....
Power Corrupts,Absolute Power Corrupts Absolutely, leaving one person(group)in charge is absolutely corrupt.
"the reason for difference in price" - testing cost.
;-)
Think on how ridiculous this is:
No-one in their right minds deploys business-critical storage with anything less than some sort of RAID protection, where the failure of a single drive is no big deal.
Customers purchasing IDE drives, i.e. home users, small biz, are much more likely to have no protection, and as such lose everything if the drive breaks!
Think about it
This is WD trying to poach the other guy's market. The traditionally produce low end, low margin drives. They envy the "enterprise classe" drives profit margins. But they know that they can't muscle in with a face-to-face competitor, so thy are tring to get into the gap between traditional "desktop" and "enterprise" market places. Might work.
Consciousness is an illusion caused by an excess of self consciousness.
>Our division manager has announced it's his goal to handle 95% of support with people reading off of a basic script with no training.
Why not simply supply the support script in the box with the drive, and let the end users read it themselves (with no training).
What's the point of paying someone to sit at the end of the phone to read what it says on a piece of paper.
No wonder the tech economy is so messed up.
"ATA disks are cheaper to manufacture than SCSI or Fibre Channel drives for several reasons. The main reason is that ATA disks are tested in batches, whereas SCSI and Fibre Channel drives are tested individually. "
What a pile of horse crap. ATA drives are cheaper because:
- ELECTRONICS.The electronics are a LOT cheaper. The amount of custom logic to support the performance requirements and features of SCSI make the ASICs much more expensive. ($20-$30+)
- SUPPORT. The main reason SCSI/FCAL drives are so expensive is the hand-holding that the big OEMs require when integrating drives into their boxes. "I had a hard error. Fly someone out here tomorrow". Yes, if you buy a drive at Fry's, you don't get this level of support. SCSI manufacturers could care less about drives bought individually through distribution. That is the dumping ground for drives they couldn't sell to an OEM. Many of the big OEMs ship ten of thousands of drives a month. That is who these drives are being made for. There are entire teams devoted to each big OEM customer.
- CUSTOM FEATURES. This goes hand-in-hand with support. Each of the big OEMs requires custom code and electronics features. There are multiple developers per customer to make this happen.
- QUALITY. In order to keep desktop drives cheap, the manufacturing yields must be very high (90%+). This isn't done through creating superior components. It is done by shipping any component that isn't dead into the field. Crappy parts shipped = high failure rates. Don't believe MTBF numbers, they are a crock.
Now, that said, there is a move towards using desktop drives in low-end server apps. The main reason is obviously cost. Many OEMs would like to drive this into the middle and high-end ranges as well. The OEMs are under the misconception that they can get a desktop drive and that it will be supported like the server drives, have equivalent performance and reliability. Given the extremely low margins on desktop drives, this isn't going to happen. Is there any reason that desktop drives can't be made more reliable and feature rich? Of course not. But it is going to cost youAnd yes, I have a clue. I work in server-class HDD development.
I have no idea how you can relate technology maturity to faster and cheaper testing. Paint is probably the most mature technology I have ever worked with personally and the testing requirements were insanely expensive. I'm not talking about consumer house paint, mind you, but automovtive paint. The time spent testing and developing new tests and testing methodology was insane. Back when they finally developed a method to use lasers to determine application quality (at a huge R&D expense) did it shorten the time it took to inspect a car? Yup, but guess what? Now *every* car gets inspected so the time spent testing is actually greater.
You don't even want to get me started on car tires.
The problem with "mature" technology from a QA standpoint, is that your customer begins to expect the product to be perfect everytime. Remember Micropolis SCSI drives? Everyone remembers them because they were so crappy, but the equally crappy BigFoot drives have mostly been forgotten because no one expected much from an IDE drive at the time.
ATA disks are cheaper to manufacture than SCSI or Fibre Channel drives for several reasons. The main reason is that ATA disks are tested in batches, whereas SCSI and Fibre Channel drives are tested individually.
That's such a crock. I can pay about $200 for a 180GB ATA drive. I just paid over $1200 each for several 180GB SCSI drives, and that was the best price I could find.
So, they're saying that the thousand dollar difference was because my drive was individually tested? Heck, I'll revolutionize the SCSI drive market by cutting the manufacturers' costs in half by personally testing each drive at my new business for only $500 each! C'mon, it costs them $50 to test the drive.
Some of the thousand dollars goes into better parts, these are good, fast drives, but most of the difference is pure profit because they know SCSI is better, that the server market needs SCSI, that people need tons of storage, and that they can collude to get those prices.
Yes, I do think the FTC ought to check into it.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)