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."
"the reason for difference in price" - testing cost.
That was indeed the most cedible information I have ever read in the ATA/SCSI flame-war.
Also, there seems to be a five year warranty coming up on the Serial-ATA from Western Digital!!!
So, what are the options for a home user, who wants to buy a reliable hard drive? I know three people who have had hard drives fail in the last 2 years. This looks like an option, but a fairly expensive one (comparatively - if I'd just fallen through a time warp from 2 years ago, I'd be out there buying one now).
being IN the semiconductor test industry, it's really interesting how rarely does people really consider the necessity, and challenges, let alone costs, in testing.
few people realize that, for example (I am saying this example purely based on speculation, but a well-formed one) that the athlon MP chip cost difference is in a large part the extra test they run on it. You see - testing cost money, anything that would make test run longer means that more money has been spent on that part "making" it. One of the things the test industry is always talking about is speeding up testing, as a way to reduce testing costs.
aaanyway... next time anybody look at some nifty / advanced gadget, think to yourself "how the heck do they test THAT?" especially with things that have fast interfaces or embedded components...
anyway. erm - to stay on topic: ATA drives could handle 10k platters; I think the point about scsi has always been the more "industrial scalability / reliability / throughput / whatever" that's the selling point. well, and the fact that back in the day you can't buy IDE CDR drives.
My life in the land of the rising sun.
So ATA's are tested in batches while SCSI's are tested individually. I think I will continue to run my business critical DB's on SCSI in that case. I just don't think I could sleep at night knowing I trusted some muppet at the factory to pay attention to the test result stats and report them back to the designers/production guys correctly.
"I kill you! You no good 56'ing!"
i've gone through so many server scsi disks here that were really expensive. seems like the quality really isn't any better than ide.
all you want to avoid is getting rid of your information that is stored on the disks. any responsible it-manager will buy raid systems so it doesn't really matter if you pop a broken scsi or ide disc out of the array and replace it.
i don't see any point in buying scsi with expensive discs, expensive controllers and expensive cables.
We are all individualists!
Bingo!
Let's see you have 14 drives on a single IDE chain and then do a copy between drives.
Or how about the simple fact that you can get SCSI Ultra 360 that are nearly 3 times faster than anything you can buy that is IDE.
Or the fact that My SCSI drives come with 5 year warranty's The only SCSI drive I have ever had fail are reallllllly old. and EVERY scsi drive I have in service (over 120 of them) haven't been spun down or sat idle for over 4 years.
The new IDE might be close, but until they get proof of reliability under their belt like SCSI has It's only a watch and see item.
SCSI is known to be bullet proof and faster. enterprise ATA is not. so the next 5 years they had better not pull an IBM and produce the worlds crappiest drives.
Do not look at laser with remaining good eye.
The biggest SCSI drives I've seen are just less than 150Gb but Maxtor makes a 250Gb ATA drive. Is there a technical reason why there isn't size parity?
I've had a preference for SCSI drives for years and I've come to accept that I have to pay a steep premium (and now I know why) but what frustrates me is the density, or lack thereof, with SCSI drives.
I don't entirely understand why it is a 36.7GB drive? By this I mean, why do SCSI drives usually go up in multiples of 9GB (i.e. 9, 18, 36, 72) whereas IDE hard drives tend to go up in 10's, etc. (at least recently)? And since this is IDE, why does it have a size more akin to that of a SCSI drive?
Thanks,
Behlal
When it comes to concurrent access, which basically means "busy server", ATA just doesn't cut it. ;), or just plain lie.
We had some entry-level Sun (netra X1) with IDE drives collapse under medium load, just because of logging. I've had older, slower, SCSI suns perform under much more load without this kind of issue.
ATA is ok for hoarding pr0n, it's OK for the live backup system; but I'm not putting those into any kind of serious server.
And don't you mention ATA RAID. Those who do never used real SCSI Raid (as in "Enterprise" RAID
It's a cost/performance tradeoff all right.
ATA had many uses, but stops short of anything inside a 19" rack.
In my opinion, Scientology is a cult you should avoid.
SCSI is faster and more flexible, and perfect for use in the enterprise. Remember, "enterprise" = high profit margins.
SCSI drives tested individually? Of course, they are meant for enterprise use, blah, blah! But if that is the case, why aren't enterprise ATA drives not tested individually too, eh?
I am sure the extra testing made on SCSI drives puts the price up, but is that necessary? Why not just mass-produce them like ATA?
Mass produce SCSI, and it will kick ATA's butt all around the room. Hard drives manufacturers just want to hold on to their enterprise cash cow by keeping production down to low levels, and keeping margins high.
Uhm, no. Read the article. The drive has a capacity of 36 GB. So the data tranfer rate will be slower, compared to a current high-capacity drive. Server drives are optimized for access time, not transfer rate. That's the reason why they keep increasing the rotational speed, at the cost of data density: rotational latency (the time a R/W head has to wait until a certain sector passes underneath it after it was positioned above the right track) is decreased.
Higher transfer rates are reached by putting multiple drives in a RAID configuration. That's also the reason why you'll not see any benefit from putting a single server drive in your desktop PC.
Did you know you can fertilize your lawn with used motor oil?
Both the Xserve and Xserve RAID use ATA drives. Why wouldn't they benefit from faster ATA?
Or just buy yourself an inexpensive 3Ware IDE RAID controller. The RAID-1 two disk controller is only about $120. The RAID-5 supporting 4 channel one is around $400 if I remember right. Considering they have built in Linux support and they have open source drivers we should really help support this company because companies like this are few and far between. I took my RAID-1 controller out of the box, popped it in my new system, put two "special edition" WD 80GB hard drives on it, created the mirror in the card's setup, and booted Linux and it recognized it as a SCSI controller with a SCSI disk attached. Just seems like a cleaner solution than mucking around with software raid. I never did like software raid much.
Memory is 10x more reliable and more shock resistant. It is also nanoseconds rather than milliseconds and doesn't take NEARLY the power or the "pixie dust" to produce.
A company called ADTRON makes SCSI 2.5" Flash drives. I bought one used on eBay (1 gig) about three weeks ago. I put it in a PowerBook Duo (1995 laptop). The Duo now lasts as long as a modern iBook and the difference is between night and day in App launch, speed, and most unforseen, graphics display. It appeared as if I had almost doubled processor speed.
If you want to see if I'm telling the truth. Look for SCSI Flash or IDE Flash 2.5" drives on eBay and try it in your laptop for a day. There are regularly 350 meg IDE laptop drives for sale. Right now the capacities are capped at 4 gig (and the price on one is $4600) But if WD, Seagate, Maxtor and all the other platter people would just get with the program I'm sure we could have MUCH smaller drives than current systems, with much denser capacities than even today.
I don't see why laptop manufacturers don't push this very hard. Battery life is almost doubled (no moving parts) and it almost eliminates the bottleneck that laptop hard drives have. As for desktops, you could have 4 of these drives in the space of one and possibly have them raided!
Yell & scream & rant & rave... it's no use... you need a shaaaave ~ Bugs Bunny
ATA disks are cheaper to manufacture than SCSI or Fibre Channel drives for several reasons.
One really huge difference lies also in electronics. Usually it's called SCSI Control Blocks (short: SCB's). They are actually commands, sent from SCSI controller to devices telling them what to do (read or write data, etc.)
Any decent SCSI drive will support at least 32 of them and it will execute them out of order, mostly optimizing head-movements. Which gives huge performance boost under truely multi-tasking system.
That is not the point. It is not that Scsi is significantly better as an interface, it is that Scsi has become a "marker" for drives manufactured to a higher spec, which need a higher price. It is more like "business class" in flying - you travel in the same aircraft, but a get a better class of service. it is not in the self-interest of any manufacturer who sells in both marketplaces to devalue the premium-grade scsi marker. Howeve, since WD sells only in the "budget" IDE marketplace, they have an interest in upgrading the image of their brand. What we may be seeing here is the start of the breakdown of an ad-hoc, non-conspiracy-based, "industry standard"
Consciousness is an illusion caused by an excess of self consciousness.
"SCSI is known to be bullet proof"
That's just plain wrong. I've had batches of SCSI drives with high failure rates. When the maker screws up the glue and heads start falling off, scsi versus ide doesn't enter into it. SCSI cabling, termination and a shared bus can also be problematic as can subtle diffs in SCSI protocol implementation.
The drawbacks of IDE have historically been: not offered on the 10-15krpm drives, cruddy cables, can't do >1 drive per channel, many broken implementations, lower qa standards.
Oldtime drives had no digital hardware onboard. It made sense to integrate things to the point where the device can locate its own sectors, but it's arguable that SCSI puts too much on the drive. I'm in favor of the 3ware Escalade style architecture where each drive has an independent channel, and is treated as a relatively dumb device.
With the improved cabling, qa and spindle speeds, I think we're about to see some really rockin' IDE storage systems.
I remember reading a couple of days ago about DRM-specific instructions being included in the ATA standard. So, does the SCSI standard have any DRM-like instructions? If not, anybody else see this as a way of getting DRM into the enterprise market? Last place I want DRM hardware is MY server!