Slashdot Mirror
ATA133 Controllers Have Arrived
Spot writes "If you're a hardware junkie, then you may already know ATA133 is on it's way to becoming the new standard for drive controllers. LittleWhiteDog has a very detailed look into the Promise Ultra133 TX2 Controller and Maxtor's D740X-6L ATA133 interface drive. " And I just bought a few 100g drives :) I still find it funny that every couple years I buy new hard drives always for around $200... 120 megs, 800 megs, 2.5G, 12G, 30G, 100G. I love this.
I can't wait to get my hands on one. now only if IDE drives were faster then 7200 rpm
access time matters more to me then disk bandwith
Still rules for now, when will serial ATA will come out for the consumer market? seemed like a slick deal for me. As for ATA 133, it's just a holding tech until Serial ATA comes out (god knows when)
kawai
That is always the great thing. They just get bigger
How is the linux support for ata133 interfaces??
If I had no sense of humor, I would long ago have committed suicide. -Ghandi
Why did I but that Ultra160 anyway?
||| I still can't believe Parkay's not butter.
due to the limitations of 32bit PCI, this effectively makes my ultra160 SCSI controller and hard drives obsolete as far as transfer speeds are concerned (and i assume the ata133 will be considerably cheaper than u160 devices of the same size based on past experience with IDE vs. SCSI). well, 15,000rpm is still nice though :)
anyone have good reason now (other than slightly superior seek times) to stay with SCSI solutions?
http://saveie6.com/
I still find it funny that every couple years I buy new hard drives always for around $200... 120 megs, 800 megs, 2.5G, 12G, 30G, 100G. I love this.
:)
:))
You should see what kind of drives they are just *giving* away these days...
2 gigs? I'll never fill that up!
(seriously, you'd be suprised how many people consider their old 2 gig drives to be in the same league as their old 30 meggers a few years ago!
It's been a long time.
*sigh*
and I've had my new ATA100 60GB hard drive for a week and a half.....
But they aren't doing anything to make it SUCK LESS. Drive platters aren't getting faster at the rate the controller is. Very few, if any, drives currently available can saturate an ATA33 bus, sustained. The only thing these ludicrous improvments are doing are increasing performance to and from the drive cache.
Now that IDE has for all intents and purposes killed SCSI on the desktop, you'd think that they'd expend a little fucking engineering effort to make it so that you can control more than two drives on a controller, and so that a other devices on the chain can work while one is processing a command.
I'm horrified at how IDE has flourished. It's the worst possible standard for a drive interface.
SCSI is dead.
For most consumer and single-user environments, IDE is plenty fast enough. Even in the small server market, IDE is adequate. In the high-end server market, people are moving away from SCSI in favor of Fibre Channel.
IDE is squeezing SCSI out of the low end, while Firbre Channel is doing the same to SCSI in the high end. SCSI won't be around as a serious disk option for much longer, I suspect.
(Not to mention that USB has killed SCSI for things like scanners.)
are commands sent over the IDE bus synchronous? i remember reading a few years ago that one of the major differences between SCSI and IDE was that SCSI controllers could take commands out of sequence. anyone know anything about this?
Given the current speed of IDE hard drivers - ATA 66 is overkill let alone ATA 133. Hell, ATA 33 is overkill for all but the fastest drives out there. The only benifit you will see, is that the drives onboard RAM-chip cache can be accesses quicker, and that moving from an older IDE spec will get you the new fangled sheiled cables that may help with reliability.
Moneyed corporations, non-working 'poor' and criminal prisoners are turning productive citizens into tax-slaves.
So what's the deal, is ATA133 a viable alternative to SCSI in low end (1-2 CPU) servers? Does it play nice even at high loads? Will I get decent performance even when all drives are accessed all the time (RAID 0+1 on a busy NFS server)? Does it support hotswapping?
And when is SerialATA due? Those stiff cables aren't any fun at all.
READY.
#
My friends "in the know" say that all this ide-spinoff stuff still suffers from the major drawback that it uses a lot more cpu than scsi, and hence the bandwidth numbers they proclaim are not achievable on a standard system...does this new ata stuff attempt to address this, or have I been trolled in the first place?
Slashdot 's editors are dickheads
And I just bought a few 100g drives
Maybe they will give consumers a bulk discount when buying by the kilo.
To-do List: Receive telemarketing call during a tornado warning. Check.
We just got some Serial ATA boards and Drives where I work, and I think that anything else is just a joke. It will hit consumer boards soon and then the real increase in speed through the system will get the boost, until then I'm just going to leave my ata33 system as is.
I moved my system from a 5400RPM ata/66 to a 7200RPM ata/133, and I have noticed a quite dramatic speed improvement. In the limited benchmarks that I have done, I get about 40MB a second unbuffered read from the drive. This is about double the speed that I got from my old setup. On a more subjective level, compiling and development with a sourcecode base on the new drive as opposed the the old one is much more responsive (I still have the old drive hooked up).
Using Then and Than.
Then: Back then, we made ten cents an hour.
Than: I can eat more jello than you.
Then: Go ahead and leave then!
Than: Better safe than sorry.
"Would it kill you to put down the toilet seat?" -- Maya Angelou
Seems that IDE/EIDE drives are the choice for cheap and large. I'm certainly guilty of buying a few. However, I am wondering why fibre channel and SCSI aren't more popular for the desktop?
/. a few months ago about an interface gadget that let's you chain them with CAT-5 ethernet cable. That would rock!
For application installs and OS install/cache, a 10,000 rpm LVD Ultra160 is hardly fast enough for me. Also, I have 9 drives on this system. I can only do 4 with IDE, and if I put in a second controller, I blow another IRQ (of which there are only 10 available of 16 - sad commentary on PC architecture). Please feel free to correct me if I'm wrong on this.
Allow me to extoll the virtues of SCSI/LVD:
-15 drives/devices per IRQ
-Lightning fast.. 320 mbyte/sec now
-Doesn't slow down your CPU when moving or copying files from one drive to another
-The above applies to burning CDs as well (a major bonus)
Basically, with all this going for it, why isn't SCSI more popular (and less expensive)?
And what about fiber channel? Seems there was a story on
Why is everyone buying IDE? Or are they? Just curious.
Vortran out
Knowledge is like ignorance.. too much can be just as bad as not enough.
Maxtor's press release from Oct.29 is here, and contains this piece of info:
:)
Ultra133 TX2 increases data transfer rates between a hard disk drive and a personal computer up to 33 percent compared with Ultra ATA/100 controllers [...]
Duh. I suppose maxtor's 160Gb drive increases hard drive capacity with... well... up to 60% compared to 100Gb hard drives also.
Ultra 100 controllers are typically moving data at less than 1/3 of their rated capacity from almost any modern ATAPI drive. As the article says: In the speed arena, the added bandwidth an ATA133 compatible controller can give you is unfortunately not a selling point at this time. I always get a kick out of people replacing their Ultra 66 controller with an Ultra 100. They are invariably disappointed by the almost identical performance. Now everyone with Ultra 100 controllers can rush out and buy Ultra 133 controllers and experience that same disappointment all over again.
What Ultra 133 buys us is the ability to use drives in excess of 137GB. Suddenly, 160MB drives are showing up that use this new standard. And that's a lot of p0rn!
Now what I want is a drive standard that can support high speed, multiple drives (not just two) per channel, is low cost, and uses a better, more convenient, round cabling system (e.g. fiber, coax, etc.).
-henrik
Imagine a RAID of those :)
----- Whats wrong with this picture? http://www.revoh.org:1234/whatswrong
You can get information on Serial ATA at serialata.org. You will find that these new ATA controllers break the 4 drive limitation, and have a very small cable, as opposed to the air-flow-blocking current ATA cables.
Another mini-rant I have to get out of the way, is about the psychotic SCSI user blaming ATA for keeping SCSI from becoming a real force in desktop computing.
Guess what, if the SCSI manufacturers would have brought the price down to reasonable levels, this would not have happened. Is SCSI better? In servers, heck yes. On the desktop? No, not really. Even on small servers, the advantages do not outweigh the extra cost of SCSI. The folks in the SCSI industry made a concious decision to stay in the server. Price DOES matter on desktops, and there is NO technology that can beat ATA for price/performance. Thats what ATA is for. Bleating that its' "technically inferior to SCSI" is stupid. They are not intened to do the same things. SCSI=Server Fibre Channel=Server ATA/Serial ATA=Desktop
i buy ide because scsi drives are too expensive. and they are expensive because i dont buy them? ;)
Imagine a beowulf cluster of these!
I still find it funny that every couple years I buy new hard drives always for around $200... 120 megs, 800 megs, 2.5G, 12G, 30G, 100G.
This is very similar to the old maxim that the computer you really want is always $5000. Only now, for that money you get a 21" flat panel display, multiple GIGAhertz and more RAM than you can shake a stick at.
Supposed to ship today, THAT'S the baby, raid-5 with 48bits LBA support. That means 960GB (6x160Gig using maxtor 160GB drives) of storage for dirt cheap, plus Raid-5 support.
:)
I am planning a non-critical datacenter (rendered frames and so on) with that setup, it's crazy, while a single drive is not offering the performance of the barracuda 180GB 7200rpm drive from seagate, it's like C$500 for a 160GB drive whereas the seagate would cost me around C$2500, you can get to the same performance (plus increased storage and safety with Raid 1 or 5) for the same price than a single seagate drive. it ROCKS.
I can't beleive I payed C$300 for a 40MB on my amiga1200 not even 10 years ago
--- Metamoderating abusive downgraders since my 300th post.
From the article: "The TX2 is the first Ultra ATA133 controller card that has support for 66MHz PCI motherboards (32-bit @ 66MHz as opposed to the current 32-bit @ 33MHz - not the same as 64-bit @ 33MHz). Granted there are no 32-bit 66MHz PCI motherboards available at this point in time (they'll be here "when they're done") but when they are available this card will be able to take advantage of the extra hertz."
It seems that we have two competing PCI slot standards - 64-bit/33MHz and 32-bit/66MHz. I assume that eventually we will see 64-bit/66MHz.
I remember an article from a few years ago talking about what the next step in PCI slots would be, and it spoke to these two steps. The argument against 64-bit slots was that it would have to change the physical dimensions of the slot to accomodate the additional bits being passed. The problem with 66MHz slots was cross-talk and RF interference between two adjacent slots.
Since these new ATA/133 cards are backwards compatible with 33MHz slots, I must assume they found a way to reduce RF interference. The existence of 64-bit PCI slots means that industry has found a way to move 64-bits using the older physical architecture.
That said, which of the standards do Slashdot readers think will catch on? Or will the two compete until a 64-bit/66Mhz standard is agreed upon?
- (c) 2018 Hank Zimmerman
I can answer two of your questions.....
With new processors, how much are you REALLY giving up in processor useage? This was only a problem on Pentium and 486 processors.
On the Fibre Channel front, FC is used for external disks. FC has a maximum distance of, someone correct me if I misremember, 2 kilometers, on optical fiber. The controllers are very expensive. The drives are expensive. The entire point of FC was to get over the 15 drive limit of SCSI and to get over the distance limitations of SCSI (3 meters) and Diff. SCSI (15 meters).
I am not aware of any internal FC implementations on standard server hardware, but as a rule, its an external JBOD application.
How utterly fascinating and insightful.
--This isn't a man who is leaving with his head between his legs.
It seems to be consipciously absent from most motherboards and hard drive. It sounds great in practice (a dedicated channel for each device, faster transfers, smaller footprint) but Intel has explicitly declared that their chipsets will not support it. I guess it would require a brand new chipset, but still, it would be a benefit for everyone.
I just can't see the rationale for using ATA-133 in anything. ATA as a server interface is generally a bad thing unless done VERY carefully. SCSI has transfer rates that are up there (I think differential SCSI has a 160MB/sec transfer rate, and the drives are like twice as fast seeking as ATA drives.) and the drives are generally more reliable, or failing that, eaiser to replace. The average home user has no need for anything above ATA-66 or maybe ATA-100.
Because it's cheap.
Like someone said up above, it's because the SCSI vendors decided to stay in the Servers that the price never came down.
Quick scan on Pricescan.com
Cheapest large SCSI drive there
Seagate Cheetah 73.4GB 10K Ultra160 SCA
$635.00
Cheapest medium SCSI drive there
BM Ultrastar 36LP 36GB 7200 Ultra160 LVD
$210.00
For ATA-133
Maxtor DiamondMax Plus D740X 80.0GB Ultra ATA/133
$195.00
I know that SCSI is better. But is it worth getting the SCSI card and paying alot more? Not to me it's not. I play some games, mess around on the Internet and thats it...SCSI won't make that any faster.
As cheap as I've seen fibre channel controllers on ebay, or even the wholesale outlets, you _MUST_ be nearly braindead to continue to go with IDE. I don't care how they've managed to eke a little more performance out of it...
Seriously, it wasn't 6 months ago that slashdot itself was exalting cheap fc cabling systems ( http://master-www.cinonic.com:8080/ ). This, wholesale $25 nine gig drives (small enough that they're useless to corporate SANs, maybe up to 30 or so gigs for people like us), and a maximum device limit of 128 drives says it all. 200mps, and even with software raid, you'll be able to approach that speed. Think about it. Raid 5, no more hd crashes that wipe you out. Speedier than all but the most exotic scsi setups. You can spend $50 week, and incrementally add storage. Honest to god, we should be shunning hardware manufacturers that continue to feed us this slop called ata133. They're the same ones to blame for marginallizing scsi, and making it more expensive than it needs to be. Don't let them do the same thing this time around.
Moderators, I dare you to mark this as flamebait or trolling...
Geez, compare apples to oranges much?
Using the Cheetah 73G SCA drive as a comparison to an ATA-133 drive is so misleading it's not funny.
Or are you implying that you can hot-plug that DiamondMax?
... and maybe PC. I noticed that a hardware store around here in the UK has a completely different ATA133 product, or two. www.pc500.net has something by ACARD. They're advertising it as an ATA133 card for Mac with RAID 0 striping. Dunno about the PC, but will these work in the same way as the one's that come with the drives? Also, will the PCI cards from maxtor,etc. work with other platforms than PC's?
Basically, with all this going for it, why isn't SCSI more popular (and less expensive)?
SCSI isn't more popular precisely because it is so damn expensive. To use your example, who exactly can afford the price of an Ultra160 controller and drive just for "application installs and OS install/cache" ?? Precious few people. Who *really* needs more than 4 drives - very few people, especially when bigger and bigger drives just get cheaper and cheaper.
And SCSI isn't cheaper because there is the less expensive IDE always available. Even if SCSI could be made as cheaply as IDE, good marketing people will always price SCSI devices more than comparably-sized IDE devices, because the people who need SCSI's features are willing to pay a premium over IDE. At least in this regard, it is all about market segmentation and differentiation.
Trust it to the french to build a plane where the engine falls off -- that's it FALL RIGHT OFF -- on takeoff.
Now slashdot is playing into their hands by advocating their cheapo-french CPUs.
What's next? A tricolor flag on the white house?
dude, whoever rated the above poster a Troll is abusing their moderator power (that never happens on /.). "i am not a troll! i am a free man!"
Even the fastest ide harddrives can only send data at a constant stream at a max of 35 megs/sec. What is the point of using 66/100/133? Yeah you can read the cache data faster, but the times for reading the cache for these interfaces in seconds .03/.02/0.15 so is there any significant improvement here? Hell no! What the drive makers need to start doing is making drives that can read/record data faster rather than creating faster/expensive interfaces that we don't need and the only reason anyone would get one of these controllers is that they are an idiot or bragging rights. But what real preformance are you getting out of this drive. There is no gain in using these drives, and they are more expensive, if you have the money just get a bigger drive.
"If you commit sodomy they'll put you in jail with a guy who will sodomize you." -- George Carlin
I personally have a motherboard with the ATA100 built onto it, and to be honest, it has been less than perfect. Half the programs in Windows 2000 have problems figuring out where the hard drive is, in windows xp this problem seems to have disappeared for me. I couldn't even install the operating system on it, I had to stick it on the regular channel then switch it over.
This is also a Promise controller, not some no-name where you might expect such things for it being integrated onto the motherboard.
Hopefully if ATA133 gets implemented onto some motherboards they'll figure out how to make it run a little more smoothly.
Last I heard IBM/Seagate/WD wern't going to support 133 ata...
I don't think anything has changed...
Looking at the specs on the linked article:
New Ultra ATA interface with Maxtor-patented Ultra ATA/133 protocol supporting burst data transfer rates of 133MB/s.
Maxtor-patented? I hope this is a typo or editing mistake. Looking around at http://www.uspto.gov/ doesn't reveal much, but Googling for information brings up a few press releases saying things such as "Ultra ATA/133 Is Based on Maxtor Patented ATA Technology" and "The Fast Drives specification and licensing rights for Ultra ATA/133 are available from Maxtor under non-disclosure."
Are other ATA standards patented like this, by Maxtor or other companies like Western Digital or Seagate?
Ian
> I am not aware of any internal FC implementations on standard server hardware, but as a rule, its an external JBOD application.
;-)
* SunBlade 1000 uses FCAL internally (workstation)
* SunFire 280R, v880, and 3500 use FVAL internally (servers)
More and more, FC is not used for JBOD, for RAIDed disks. And more and more prevalent is the use of FC Hubs/Switches to make managed of disks much easier (in a large server environment). FC is *almost* a necessity in (failover) cluster environments of more than two systems. BTW... FC uses the SCSI 3 protocal over copper/fiber.
Todays huge harddisks don't make me store more pr0n on them. They let me store the same amount of pr0n but in much better quality! :)
My other account has a 3-digit UID.
...to spell correctly. It's PLEBEIAN - And you, sir, are plebeian.
not until i get it's ATA 166, or anything above 133. It sux that i finally get all my drives at ATA 100 and ATA 133 is out. No way, i'm not dishing out cash until the next upgrade.
I've always wondered why iterations of the ATA standard are always optimized for speed. As fast as this is, it's still only two devices per channel! Hell, even Serial ATA is still stuck with this two-device limit. Is this just a way to keep the overall costs down for the OEMs? (That is, optimize for speed only?) Or is there a technical/signaling reason why this limitation exists?
I can't wait until they finally work SCSI-like device support into the ATA specification. (Hear that, Adaptec?) That would rule. An integrated ATA133 chip with on-chip RAID would be killer.
Put two very fast hard drives on the same channel and you can push 100 or even 133 MB/sec pretty easily. Sure, it's going to be power-user and (once the RAID version of the card hits the streets) low-end server territory, but that's exactly Promise's market.
- You caught the misuse of then, but not the missing apostrophe in cant.
- You omitted the period in your abbreviation of versus.
- All five lines of your corrective post are sentence fragments; none are grammatically correct.
- The Jello® trademark is a proper noun; it should be capitalized.
Also, you didn't seem to notice that the original poster didn't even use punctuation in two out of three sentences. Obviously, it was a troll for anal grammarians. Based on your previous post, I'd say the trap has sprung, and caught us both.Don't be an asshole. If you claim that IDE transfers don't eat a lot of CPU once you actually start to stress the drive subsystem then you're either criminally underinformed, or lying.
There are a few designs, such as Sun and Intel, who are adding FC to the interior in place of SCSI, but generally, no, SCSI dominates the server internal connect. BTW, LVD SCSI is 25m point-to-point. 15 meters mulitdrop.
You can run SCSI several dozen meters, 75?, with repeaters. FC, with FCIP and iSCSI, is getting almost infinite in it's interconnect.
FC is very expensive from a card standpoint, but it is mainly intended for multibox connections. SCSI generally was limited to two servers connecting to the same drives, FC removes this limitation. You can have several dozen systems talking directly to the same disks.
They are the same. ATA and SCSI drives, and have been for very long. It's only when you get to higher rotation speeds they differ.
Making it hotswapable is only a matter of designing an interface that will disconnect all wires at the same time, and plugging them back in also at the same time. You could build an IDE drive bay that would be hotswapable.
first, boot the linux kernel with the IDE-Bus set to 66 (set the idebus=66 option), if your motherboard and drive controller supports it.
ATA/66, Non-CD, has DMA support:
/sbin/hdparm -d1 -X66 -c1 -u1 /dev/hda
Older drives, not ATA/66, but with DMA support:
/sbin/hdparm -d1 -X34 -c1 -u1 /dev/hda
The burner doesn't support DMA:
/sbin/hdparm -d0 -c1 -u1 /dev/hdc
man hdparm for more info.
"If you're a hardware junkie, then you may already know ATA133 is on it's way to becoming the new standard for drive controllers."
/. frontpage.)
ARGGGHHHHHH IT'S "ITS" NOT "IT'S" YOU FUCKING MORON.
(standard disclaimer for Mr. Offtopic: go fuck yourself,I'm sick and tired of seeing lazy writing on the
The SCA drive costs no more than the 68-pin version of the drive, generally.
- A.P.
"Remember when the U.S. had a drug problem, and then we declared a War On Drugs, and now you can't buy drugs anymore?"
FC uses the SCSI 3 command set. This allows the OS to use the same route for FC as for SCSI, which makes OS adoption easier. Microsoft is looking to re-write their stack for FC drives to make them faster though. FC RAID is very popular, but the RAID is done in the external boxes. Almost all external RAID boxes, EMC and such, are FC interconnects.
FC can also run TCP/IP and other transport forms.
By the time they release an IDE drive that hits 1 petabyte, ATA-133 will be long-dead.
"Remember when the U.S. had a drug problem, and then we declared a War On Drugs, and now you can't buy drugs anymore?"
10km generally. If you have the proper hardware (think DWDM) you can go further. Latency is the issue.
And your post saddens me, as you have completely missed the point of his post.
You're a moron. Move along now. Nothing to see here.
This doesn't mean that drives will be any faster......my IBM 60GXP (one of the fastest IDE drives available) wouldn't even max out an ATA66 interface......
I think you will also find that FC is going to be the disk-of-choice for enterprise systems in the future, but not for the interconnect. iSCSI over Ethernet will replace FC as the interconnect. Especially when you consider 10gb Ethernet that will hit the market next year.
FC is good tech, but it's hard to use, interoperability is a HUGE issue and the familiarity of server admins with Ethernet will make them inclined to use Ethernet.
Don't waste your money on this level of technology. I'm waiting for the Serial ATA to come out next year!
One reason, which others have hit on, is that it's nothing more than an ego-match with SCSI's 160 MB/sec bus speed. However, there is a semi-valid reason: The spec includes a addressing extension which increases the maximum size of a drive into the petabyte range.
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.
Prove to the owner of this company (warning: flash-based) that SCSI is faster than IDE. He claims that 'Superdell's Personal Model' is the absolute fastest Windows-based machine on the planet and it has top of the line parts, and if you can tell him how to make it better, he'll pay you $1000. Click on the right arrow twice where it says 'Dang Fast P4 model' and then click 'More Info'.
Okay. Whatever. Even in his highest-end servers he uses a RAID setup with IDE.
I'd sure like to burn this guy out of his $1K. He's really obnoxious; he insults the other two competitors in town as well as Dell, Sony, etc., and, in an act of either supreme arrogance or stupidity, he put up a HUGE BILLBOARD proclaiming that he 'beat the dentist and the Better Business Beauru'.
Seems that IDE/EIDE drives are the choice for cheap and large. I'm certainly guilty of buying a few. However, I am wondering why fibre channel and SCSI aren't more popular for the desktop?
Seems that you answered your own question there. IDE/EIDE drives are the choice for cheap and large, while SCSI are not. And cheap and large is what is necessary to store gigabytes of audio and video, unless you're wealthy. Very few of us care enough about the extra speed to justify paying $500 for an 80-gig SCSI drive when you can get the same thing in an IDE flavor for $200.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
What's really perverse, of course, is if you run:
iSCSI over IPFC over FCIP...
I suppose you could infinitely virtualize the FC/IP pairs in this scenario...
General Relativity: Space-time tells matter where to go; Matter tells space-time what shape to be.
Funny how YOU are the one who comes off as the asshole and not him... hmmm
Now my IDE drive interface can be A FULL ORDER OF MAGNITUDE faster than any IDE drives I can buy.
How about updating the spec to allow more than two drives per channel? Then we might actually have a chance of saturating it.
By the time they release a CPU that hits 1 GHz, x86 will be long-dead.
Anybody would have agreed to that 1985!
Except that ATA133 is a singlevendor spec and will be dead as soon as serial-ATA ships next year.
There's no way I'm going to fit >4000 MP3s onto a 600MB hard drive. Or one modern game.
Hardware junky and ATA don't belong in the same sentence. No hardware junky gives a shit about ATA. SCSI will always be ahead of the game
And what, pray tell, does an average consumer (WinXP user) need with fibre channel? Or more than one drive, for that matter?
Have YOU installed fibre channel on your system?
Should my 70-year old grandma put fibre channel in her computer?
You really should read up a bit on fibre channel before you go saying it is an acceptable replacement for cheap IDE.
Do normal users need "Connectivity over several kilometers" or more "flexible topologies" than a few drives daisy-chained together?
The simple fact is, fibre channel is only practical for high-end machines. If you run a small server or a desktop system, you'd be "braindead" to switch your system to a totally incompatible storage architecture, abandoning the compatibility and ease of IDE for the average user.
You also have to take into account that the cables for fibre channel are much more fragile than IDE ribbons, though not subject to RF reception in the medium itself.
For the record, I've never had a problem with IDE on any of my systems. It's cheaper to upgrade than my SCSI systems, and the drives are getting bigger and faster all the time, and not really going up in price.
Firewire hard drives, now THAT looks promising for desktop use...
"Alcohol, Tobacco, Firearms, and Explosives" should be a convenience store, not a government agency.
The drives are dirt cheap.
The controllers are cheap (not cheap as dirt, but cheap all the same).
Don't like those particular brands? Pick some others.
Remeber the flame wars polls a month or two ago
I think I want to change my vote SCSI vs. IDE Rocks my world
- 800,000 hours mean time between failure (MTBF) in the field
- 3 Year Limited Warranty
units 800000hours years* 91.263642
If their drives have a 91 year mean time to faulure, it would be pretty cheap for them to give a 5 year warranty rather than a 3 year warranty. Even if their MTBF was off by an order of magnitude , a 5 year warranty wouldn't be that bad.
I think it's time for someone to compile some failure stats on these things.
(anecdote)
Back in the early '80s when oil sands development was starting in Northern Alberta, a friend of mine was working at the site. It was mid-winter, and starting to get pretty cold... -35C (~-30F)
-30 is cold on any scale, but the equipment that they were using was rated doen wo -40. Now in the States, +40C ~ -40C is often referred to as "Mil Spec". In Northern Alberta it's referred to as "outdoor equipment".
-35C, and this equipment freezes. My friend Dan calls the manufacturer of this stuff and he complains about it. The engineer led off with one question that told Dan all he needed to know.
With a Texan drawl he asked, incredulously: "You mean it actually gets that cold?"
I'm wondering if Maxtor's 800K Hour MTBF is kinda like that Texan Mil Spec rating.
Sometimes boldness is in fashion. Sometimes only the brave will be bold.
So is this going to improve speed between two drives on the same IDE cable?
Not very familiar with Firewire, ATA and SCSI, but could Firewire replace all the other protocols? What would be the problems/limitations?
I dream of the day I buy a PC motherboard with only Firewire ports, AGP and PCI slots. No serial, parallel, PS/2, USB or IDE. Cut the costs and get hot swapable devices for anything, including HDs.
Yeah, I know that there are already some Firewire HDs selling, but there are expensive, I have not yet seen one in a mobile rack configuration and they are not standard, i.e. I can't get to my friends and simply plug my HD on their computer.
What a customer. Buying new hard drvies every other year.
I work in the operations team with in our IT dept.
We have a range of servers from high end Alphas to OS/2, NT & Linux servers with IDE.
At the end of the day in my experience SCSI uses less CPU time, has always been faster, and will always be. Yes IDE is good, we use it in a number of our servers where cost is a factor.
All our high-end database servers are SCSI, Jade anyone? SCSI will ways be better because it's technically better than IDE can ever hope to be. The only thing that will out do SCSI will either be a redesign of the conventional disk interface or a complete revolution of how disks are interfaced to the host.
We must also remember that a system or sub system is only as fast as its slowest link and to use the technology best for the job at hand.
I recently purchased that Maxtor drive, thinking I was getting an ATA/100 drive, and was pleasantly surprised to find out I'd have a drive that my controller may someday catch up with... :)
But the really great thing about this drive, is that its the single quietest drive I have seen.
Its phenominal!!!
For those of you that care about a quiet PC, I hightly recommend this drive.
How come the stores haven't put up the usual signs saying "Le Controlleures ATA-133 c'est arrive"?
Or was that joke a little too obscure for this crowd? And I don't speak French, so no bashing that...
Here's a funny quote
'Though this shouldn't be an issue with most motherboards these days, Western Digital provides a utility to force the drive into ATA-33 operation. No loss of performance would occur should this be necessary.'
from
http://www.storagereview.com/
review of the Western Digital Caviar WD205BA, which I was reading for fun.
Pretty telling, huh?
You probably like USB hard drives too.
Since you didn't read my post very well, I'll point out that it's possible to use cheap cat5 cabling. No, they aren't kilometers in length. Yes, people should have more than one drive, preferably 3, making raid 5 possible. Right there, no more need for tape backups and what not. I've read up on this very carefully, making sure that it's possible for me to undertake. I do have a small server here at home, and I'm hardly braindead. I just don't like cheap consumer junk like you apparently do. IDE? *LOL* You might have made a case for SCSI, but not IDE.
gee, try $350 for 5megs, $325 for 20megs, $300 for 40 ...
> However, I am wondering why fibre channel and SCSI aren't more popular for the desktop?
Because SCSI and fibre channel controller's don't come standard on every motherboard sold.
I don't see why there won't be a mixture of standards. IDE/ATAPI for joe consumer, SCSI for us discrimating desktop/server buyers, and FC for people who have too much money and like buzzwords.
Don't underestimate fibre channel. It's a very fast interconnect that's easy to implement over long distances with optical cabling, and that supports 127 units per loop.
I have about a terabyte of FC disk in my lab on various FC loops, and the stuff works so well it's almost funny. And our customers have, combined, many, many TB of FC JBOD. Super reliable.
And that's without even getting into switched storage fabrics. Personally I think cross-platform storage-sharing schemes like Sanergy-- and others whose names I forget-- are pretty kludgey, but shared stored in a single-OS environment works really well.
And even without shared storage, the ability to put all your storage-- both disk and tape-- and all your computers on a big FC switch and dynamically move devices from machine to machine just by unmounting over there and mounting over here... well, that's just plain cool.
The simple fact that I was trying to point out is, that ATA133 is NOT designed for geeks, it is designed for the average (Windows) user. You are not the average user. The average user has ZERO need for fibre channel, and probably not much need for raid 5. Did you know I still have hard drives in service that are over 10 years old? Not a single bad sector on them. I admit this might be extraordinary, but the fact is, 90% of the market, which ATA133 is aimed for, is made up of people who don't want something like fibre channel. It's not necessary, it's not backwards compatible.
Backwards compatibility IS important. Why do you think it's taken so long for M$ to remove support for DOS programs from Windows? If it wasn't strategic and earning them money, it wouldn't have been there.
You missed the point of the entire concept of ATA133... it's not meant for geeks or servers. It's meant for the rest of the world. This is why ATA133 will succeed, and fibre channel will be relegated to the background of the PC world even worse than SCSI. At least SCSI has some big backers like Apple.
"Alcohol, Tobacco, Firearms, and Explosives" should be a convenience store, not a government agency.
As someone pointed out, less than eloquently, if you are noticing no IDE lag on your box, then you're likely not doing an multi-tasking. Or at least none that is in any way disk intensive. So the problem is invisible to you, but far from 'only a problem on Pentium and 486 processors'. That is bunk.
LEXX
"Gold still represents the ultimate form of payment in the world." - Alan Greenspan, 1999
Lord forbid I suggest AOL Joe to evolve. My post was aimed that those marginally intelligent. Even you yourself qualify. 3 weeks ago, when slashdot had their "ultimate pc design" story (forget which site it was from) the morons chose ata. Why? I honestly don't know. For power users, this is a very valid choice, and I believe even a few people of the AOL mentality might benefit from it, provided that someone else in the house doesn't mind tossing a few gigs of network shares their way. It's a very elegant solution.
Just have a look at the SUN Blade 1000
if you want internal FC disks.........
The fact that the harddrives can't compensate for the speedincrease is one thing. But what about the transfer rate of the PCI bus istelf, when you own a Promise133 PCI card?
Let's look at some more calculations...
Currently the 32-Bit PCI bus runs at 33MHz.
(32/8 * 33MHz)/(1024*1024) = roughly 125MB/sec.
(byte * clocks)/(megabyte)
Now I could be wrong about this statement, but I believe that transferrate is shared by all your PCI devices...
(Onboard controllers might be a different case.)
So you won't get full usage of your new ultra-mega-superfast controller anyway.
(Yeah, I've heard of the 64-Bit PCI bus, and it should fix the problem)
My conclusion:
IDE = = cheap storage with good performance,
SCSI is for performace, and all that other hardware I can't cram on my IDE bus.
Sidenote:
Average transferspeeds on my Ultra160 HD are around 70MB/sec.
Average transferspeeds on my ATA/66 disk 20MB/sec.
But my point is is that it's great for fault tolerance also, not just distance.
as soon as you can fit 15 IDE devices on one IDE port at 160MBitPer second then I'll think you might be serious.
IDE is a toy for consumer level products only. SCSI was and has always been what you use for commercial or important data. A SCSI drive of the same speed that an IDE is is always faster. Disk copy from drive 1 to drive 2 uses far less resources, and IDE doesn't offer any bus that can exit the computer case for something important like maybe a DLT tape drive jukebox.
What IDE tape backup solutions exist that hold 40 tapes at 70Gigabytes each with a automatic changer and 20Meg data transfer rate?
give me one IDE backup solution that has an archive life approaching 100 years?
SCSI is very much alive, it's for high end. IDE is for low end.
They're two different products for two different things.
Do not look at laser with remaining good eye.
Hate to nitpick but I just ran a search on pricewatch.com and got this:
Seagate Cheetah 73.4GB 10K Ultra160
$420.00
From Drive Solutions.
But nonetheless price is a factor in SCSI drives.
Easy raid means never backing up, ever again. Never losing files to a crashed HD.
Careful there. RAID is very nice, but it's only protecting you from media failure. Continual, timely backups is still the best solution since it will help you restore from even the most worst case scenarios - don't discount it.
I gather that you're setting up a server for home use and aren't concerned with having a bullet proof recovery solution. That's fine, but don't let a RAID give you a false sense of security. You're still keeping all your eggs in one basket, albeit a sturdy basket.
Lord forbid I suggest AOL Joe to evolve.
Oh, come now, you KNOW that is asking too much. That's why Windows is still on top. People are happy with what's comfortable to them, whether or not it's the better solution.
My post was aimed that those marginally intelligent. Even you yourself qualify.
Uncle Sam and most other people pin me as better than "marginally intelligent"... my problem is that it doesn't show as well in an online forum. I have a harder time getting my points organized in this setting.
It's a very elegant solution.
On that, at least, we agree. :)
"Alcohol, Tobacco, Firearms, and Explosives" should be a convenience store, not a government agency.
Apparently he's more intelligent than you.