Slashdot Mirror
Serial ATA Technology Explained
Mike Parsons writes "Explosive Labs has an interesting article on Serial ATA . Here is a quote: 'In the rapidly moving computer industry, there are rarely the kinds of revolutionary changes like what is about to take place in secondary storage segment. Soon the hard drives and configuration methods that have existed since the origins of the personal computer will change forever. The basic IDE technology has been around for nearly twenty years. When the lifetimes of other computer components like CPUs and video are measured in months, twenty years ago seems like prehistory.'"
Think about this - how long has RS232 been defined? How long has the PC's parallel (i.e., LPT1) pin-out been defined? How long has the VGA pin-out been defined? How long has the PC keyboard pin-out and protocol been defined? A lot of things change pretty fast; a lot of things stay around forever. It all depends on whether upgrading them is worth the cost in the long run.
-Hentai [in vita non pacem est]
All of IDE's shortcomings are fixed by SCSI (except for a small degree of added complexity). SCSI hardware is more expensive, and rarely does it come built-in to motherboards.
If more people used it, it would be a cheaper solution, and would fix all of IDE's problems without re-inventing the wheel--it's a solution that, right now, works.
15k rpm scsi drives get seek times in the low three range--that's three times faster than your average 5400 rpm ide hdd.
Slashdot: Where people pretend to be twice as smart as they really are by behaving like children.
Where are my drives?
I got a shiny new SATA RAID controller on my new motherboard, now when the hell am I gonna get a couple of 80 gig cheap, fast SATA drives to put into a striped set?
huh? huh?
I don't need no instructions to know how to rock!!!!
So...why don't we use firewire? Isn't it faster than SATA? And the upcomming Firewire IEE 1394b should double the firewire speed to ~800mb/s. And let's not forget the fact that there are firewire HDD-s and other perhipeals on the market (though they are generaly external) or maybe, could this have anything to do with INTEL's desire to controll all components? I don't see the price as a limiting factor either.
Live for the present, learn from the past, and dream of the future!
The author then goes on to note that the 'roadmap' calls for the 2006 version to run at 600mb/s, which fits nicely with my roadmap to world domination in 2005. ...Ummmm, yeah, we'll see.
Although looking at the list of upcoming products and the manufactures making them, I don't doubt we'll all be useing this in a few years.
"A language that doesn't affect the way you think about programming, is not worth knowing" - Alan Perlis
...'cause that means prices will drop on hard disks that I can use.
The whole point of it is to be at least as cheap as parallel ATA, even cheaper. The connectors will be smaller and cheaper for example. It should also make system design more flexible since you won't have parallel ATA's infuriating cable length limits.
All data goes through the PCI bus...and it's bandwidth is only 133MB/sec theoretical. So, what does 400/800/anything else greater than about 100MB/sec in a media interface get you? Not much!
Ever read the actual throughput specs on a drive? The media throughput is not much more than 40MB/sec!!! Read the data sheets, people!
Add this all up and what do you get? Ripped off is about it!
Since the site is slashdotted, here are further links about Serial ATA:
Cnet
SATA and ISCSI
Intel Dev Paper
Maxtor White Paper
I can't wait to get my hands on some SATA devices. However, we're still stuck with PCI, here on the desktop end. WHEN will we finally start seeing the old original PCI spec phased out on the desktop end? Not until then will new technologies like SATA be able to shine. Bus bandwidth is everything these days.
So what does SerialATA offer that a Firewire connector on my hard-drive won't?
150 MB/sec?
"The defense of freedom requires the advance of freedom" - George W Bush
This sounds remarkably like the plugs we got for Rambus RDRAM: serial interface is better than parallel, first gen won't see real performance gains, stick with us kids, this is gonna be really good.
I see a decided lack of Sun, IBM, AMD, or HP listed in the adopters, which leads me to believe that this is much like the above. Sorry guys, I'm not riding the first wave of any new tech on my salary. I'll sit on the sidelines for awhile and see how this pans out.
Wu-Tang Name: Half-Cut Skeleton Get your own Wu-Na
>Forgive me if I sound a bit naive but wouldn't parallel be faster than serial?
Yes, but just like with memory, serial is cheap, parallel costs. Those extra wires just ain't free.
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
For the time being, IDE isn't going anywhere.
NOISE & HEAT will tend to outweigh (relatively) minor performance gains in consumer systems. (Enterprise hardware is another matter entirely)
sigh....we need to start using those annoying javascripts that make people read the article BEFORE posting.
When in doubt, parenthesize. At the very least it will let some poor schmuck bounce on the % key in vi. (Larry Wall)
Personally, I could give a rats butt about the speed. I don't want SATA so my drives go faster. I want it so I'm not having to spend twenty minutes doing finger gymnastics everytime I need to do _anything_ in my case.
Is it worth upgrading for? No, probably not. But id damn sure is worth waiting an extra few months for that next machine to save the hassel of those f'ing ribbon cables.
jello.
aka aron.
Why are they wasting their time on this ancient technology? Serial is too slow to sync my damn Palm Pilot. I can't even imagine what it would be like to try and transfer 60GB of media files over it. These companies should just accept that USB is the way of the future (no extra power required either!) and get to work on something that stands a chance of selling.
Karma: Good (despite my invention of the Karma: sig)
What on earth does the type of interface have to do with heat or noise?
Yes- but AFAIK command queuing is not implemented in a lot of the 1st generation controllers because it can break backwards compatibility with PATA software. Most vendors went for an easy upgrade path instead. Look for command queuing in the next generation of controllers.
"The defense of freedom requires the advance of freedom" - George W Bush
I don't get it ... I quite agree that, as a serial bus, it'll be clocked a lot faster than IDE ... but a simple back-of-the-envelope calculation tells us that it has to be at least 8 times as fast as the current devices (it'd have to be 533 MHz to be on par with ATA-66)
It looks like a technology whose main purpose is to make things incompatible, and thus require people upgrade more stuff. And anyway, it's not the speed of the bus the limiting factor (for the vast majority of users), but the mechanics of the harddrive (SCSI hardrives are faster than IDEs because they almost always are top-of-the-line products with higher rotational speeds - anybody saw a 15000 RPM IDE ?)
The Raven
The Raven
Absolutely. But this has nothing to do with SCSI, it has to do with the high spindle speeds at the bleeding edge. The card on the underside of the drive is not making that ear shattering racket. They even acknowledge that in your quote.
SCSI is better than ATA. Even SATA. ATA has been trying to catch up by stealing some of the best parts of SCSI (like TCQ). But it just isn't quite as good yet. Quite frankly, I agree with the majority of SCSI zealots: if the damn PC makers would embrace SCSI, then the cost of SCSI would come down to near parity from the volume of sales.
Now, is SCSI better for your average Joe? Maybe not significantly. Neither is 7200 vs 5400, 2MB vs 8MB buffers, or 8.9 vs 9.1 ms access times.
However, if they could use one cable to connect 15 devices in their tower, they'd be alot happier than having the 8 cables they'd need to do it with current IDE tech (let alone IDE's relative inability to be used externally).
The only thing I haven't seen is any noise about chip sets that support in on the system side. As soon as these are available, you'll see MBs and systems. SCSI will probably stay important for larger faster arrays, but scaling bandwidth seems to look pretty good for this as well.
As soon as mainstream MBs are there, these will quickly become the commodity drives for all the manufacturers, and they will phase out Parallel ATA stuff.
Note that almost all the Firewire hard drives that you see on the market are ATA hard drives with FirewireATA hardware in the enclosures. As far as I'm aware, the only disks that you can readily get your hands on will have interfaces of IDE/ATA, SATA, SCSI (of various connectors), and FC-AL. That's why you can't use Firewire inside a PC. Using SATA makes far more sense, especially for migrating to a new standard, as it's most likely easier to make a SATAATA adapter since the protocol is very similar.
This post is an *excellent* example of what watching too much super-fast cutting TV (like MTV) can do to you.
Note that Sun, IBM, AMD, nor HP are disk manufacturers. (Well, IBM might still be, my memory is being bad tonight, but I digress...) Some AMD and Intel motherboards are already coming with SATA RAID interfaces. Intel is right behind the technology, as they are a chipset manufacturer. AMD isn't. VIA, a chipset manufacturer is, along with a ton of other manufacturers who are core to desktop/workstation storage. Just because the big power houses that you name aren't on board doesn't mean anything. Most of these places leave their disk interfaces up to someone else. And those companies *are* adopters.
Forgive me if I sound a bit naive but wouldn't parallel be faster than serial?
On paper, parellel can be made to be faster than serial. However, in the practical world it is very difficult to make a high bandwidth parallel bus. It is even more difficult to run that bus any considerable amount of length. By using a serial bus with embedded clock you only need 2 signal wires. If those signal wires are a differential pair (perferrable Low-Voltage) then you can run them a considerable length at an extremely fast rate. If you have a parallel bus, you don't have the option of embedding the clock. If you are running with out an embedded clock you must send a clock syhncronous with the data. Now you have to deal with skew issues between each individual channel as well as all channels relative to the clock. Not to mention other aspects such as crosstalk between the channels. If all you have to worry about is two signals (which if are differential can be considered 1) then many of those issues ago away.
There's a lot of physical behind why its extremely difficult to run long lengths of parallel lines. Yes a parallel bus is faster, but it is almost impossible to implement a reliable parellel bus running at 1.5Gb/s, through cables, and with connectors. Take a look at a bus like hypertransport. It can be up to 16bit wide and run 1.6Gb/s. However, it is a point to point protocol, is run over a control impendance, and is run over very short distances.
I hope that helps.
I think I am. Really.
The article seems immensely biased and lacking in technical detail. It also raises some "dubious" points IMHO. Let's see:
- P-ATA cables cannot be longer than 40cm. S-ATA cables can be up to 1m long:
Granted, those cables are annoying. But really, how many times have you felt the need for a cable much longer than 40cm? People with full-sized cases may benefit, but then the author says that the current trend is "small footprint machines". So, why do I need a cable that is bigger than my server?
Also, if you dislike flat cables, buy "rounded" P-ATA cables (available today, just google for it).
- P-ATA connectors are big!
Yes, they are! But you'll require at least twice as many S-ATA connectors, as only one device is supported... In the end, the real state on the mobo is going to be similar.
- One device per controller is an "Advantage".
C'mon... This guy must be joking. I couldn't believe my eyes when I read it! One device per controller is an *advantage*???? Why??? I wish I could add more devices (like SCSI and Firewire) to my curreny P-ATA technology. And then he says ONE is good for me? Don't think so...
- High transfer rates are useful for multi-disk RAIDS.
What kind of RAID? RAID 5 is slow in writes due to the computational power needed to calculate the XOR. Adding bandwidth won't help. And I can't see why or how only RAIDs will benefit from higher throughput.
- Speed:
Granted. It may be faster than P-ATA. But what about established technologies like SCSI and Firewire? I *think* (not sure) Firewire can go much faster than S-ATA in its initial version.
I'm disappointed...
Of course, no discussion of Serial ATA would be complete without mentioning the answer from the SCSI camp - Serial Attached SCSI. SAS will use the same connector as SATA, but will support longer cable lengths, multiple initiators (if you don't know what an initiator is you don't even belong in this discussion), full SCSI semantics instead of lame-o ATA semantics, etc. Even so, the SAS folks are still ceding the high end to Fibre Channel and talking about three coexisting technologies for the low-end/midrange/enterprise market segments. Sorry, kiddies, but SATA is still low-end.
If there's one mistake you should try not to make more than once in this business, it's that competitors have been standing still since their previous generation. Announcing something brand new and having it be less than half a generation ahead of the competitor's last version is a failure.
Slashdot - News for Herds. Stuff that Splatters.
Firewire (1394) was killed by Apple's licensing fees and Intel's sudden backstabbing policy change on building it into south-bridge, along with their NIH attitude. There existed working 1394 Device Bay drives over 6 years ago, with OS support from m-soft. 1394 was an attempt to keep the good parts of SCSI protocol, while leaving out as much of the useless stuff as possible (MODE SELECT).
Fibre-channel is still Real Pricey, for the same reason that SCSI is -- "just because". Or, as the hardware vendors say "harrumph, well, it's all about volume".
The Russians have won. They have made the world a cesspool of distrust, greed, fear and hate.
IDE also becomes outdated very quickly.
Try plugging your 7200RPM 120Gb IDE drive into a 386 era IDE controller and see what sort of performance you get. You'll probably only be able to access 8Gb of its capacity also.
IDE hasn't "just been there" it has been constantly evolving.
First, this is IDE (just as ATA-66 or ATA-133 are IDE).
Second, the reason why Betamax died (well, didn't actually die, but didn't take off, either) was Sony kept it a proprietary format, while JVC let pretty much everyone make VHS products.
Serial ATA is one of the most unrevolutionary evolutions ever made. Basically it just changes the cables. The drives stay pretty much the same, the controllers stay pretty much the same, the drivers can stay exactly the same. Instead of wide, flat cables and two disks per channel you now get thin round cables and just one disk per channel (but since the connectors are so much smaller, you can have many on the same board). It's a good thing.
There are basically three reasons for having multiple standards. The first is a purely commercial one. Brand A invents the A-link and patents it, and brand B decides to create B-link so they don't have to pay a fee to Brand A. The second is evolution. Sometimes, a standard needs to be replaced or updated to cope with new demands (ex., ATA-33 becomes ATA-66). The third is that some standards are specifically suited to some situations (ex., SCSI lets you connect a lot of drives, and has support for other kinds of peripherals, but IDE is cheaper to make, and enough for most people).
RMN
~~~
400Mb/sec vs 150MB/sec
Pay attention to case, it does matter. As for what's in development, call me when its actually available.
Matt
Serial ATA support is in Linux 2.5.35 and up, as noted here.
It's real simple, ribbon cables SUCK, they cost more to make then serial so PC makers hate them.
So, here's how it is...
Fibre Channel - 2Mb/s(10Mb coming very soon), 126 drives, 10+ mile range, better then SCSI.
S-ATA - 1.2Mb/s(2.4Mb in 2004), 18" range?, IDE protocols for all your write-only data needs.
S-ATA is the Ghetto FibreChannel, just like IDE is crappy SCSI, expect similar suckiness and low quality to go with the low price and cheaper cables (to make, to buy they will cost more I'm sure).
But again, this is all about the creaper cables, since lets face it 95%+ of the machines out there only have one drive anyway.
- Adam L. Beberg - The Cosm Project - http://www.mithral.com/
Please, for the love of everything that's geek... DON'T use "drop" and "hard drive" in the same sentance.
...and one reason only:
To get rid of those damn ribbon cables.
Don't believe the marketing hype. SATA isn't about faster speeds, or more advanced features, or any of that crap. S-ATA is about cables.
IDE is crippleware. At some point in the past there was probably a need for a simpler, less expensive counterpart to scsi for desktop systems, but frankly that need is gone. The price distinction between IDE and SCSI has long been totally artificial. Drive manufacturers make a drive, and then slap on whatever control board they need, IDE or SCSI. Makes no difference to them, except that they get to mark up the SCSI version. Pure marketing: they need to stratisfy their technology so the enterprise guys don't feel like they're sullying their hands with the same tech as those Walmart PC-consumer lusers.
Frankly I wish SCSI had those neat little connectors (and they soon will, with Serial attached SCSI), and I hate ribbon cables as much as the next guy, but I'm not going to be fooled into thinking this is any real improvement over IDE.
But even as little as this is, it's long overdue. Those ribbon cables are the enemy of all that is good and just and true in the world.
Remember folks, SATA is only one letter away from SATAN. Q.E.D. Evil.
This is all fine and good, but why not just treat the wires in a parallel cable as individual serial wires? Sure, if you increase the signal frequency, it becomes next to impossible to guarantee that all the signals arrive at exactly the same time, but I don't see the need for bit-level synchronization. If each wire has its own protocol, its own synchronization, and its own buffers, then as long as there is synchronization at the packet level, there should be no need to worry about synchronizing at the bit level. This would allow both high frequencies, and lots of wires.
This seems to say something that I've never seen admitted about serial ATA: that it has DRM built in! If you want to buy hard drives that get to decide what you can and can't store on them, go ahead, but I'm not going to buy into any DRM technology. Extra speed and a smaller cable will not tempt me into doing it; I'll stock up on the last of the regular ATA drives as the serial ATA's replace them.
I'm an American. I love this country and the freedoms that we used to have.
Not quite. SATA controllers have one device per channel, with no master or slave.
You are probably thinking of the situation where the motherboard has a (p)ATA controller and a converter is used to connect it to SATA cables. In this situation, one SATA channel is assigned to the (p)ATA master and another to the (p)ATA slave. But from a SATA point of view, the two channels are completely independent, and only support one device each.
RMN
~~~
It, um, reads less like a press release than does the Explosive Labs piece :-).
ACARD makes a series of SCSI-IDE bridge cards which connect to the SCSI chain on one side and an ATA hard disk on the other. They have several models, mostly depending on what type of SCSI cabling you have, costing from $50 to $80. They support large ATA disks, the cost of which plus the $70 for the bridge is still cheaper than most SCSI drives. If you don't need the warranty and physical traits of SCSI hard disks, but you want to be able to hook up 6 drives to your PC with only 1 IRQ and IO address or add 60-80 gigs of space for under $200, this might do the trick. They also come in handy for old workstation-era machines, like PowerMacs, SparcStations, or VAXes. The bridge doesn't require any drivers or software to work, since it just tunnels ATAPI and makes the IDE drive look like just another SCSI disk in the chain.
http://www.acard.com/eng/product/scside.html
Microland sells them in the US:
http://www.microlandusa.com/microland/
Some downsides:
- The hard disk has to be formatted while cabled to the SCSI-IDE brige. You can't move a drive from a regular IDE controller to the SCSI-IDE bridge without getting geometry errors.
- The interface is ATAPI only, so not all commands for the device may work. FE, firmware updaters and vendor utilities designed for the hard disk probably won't work the bridge.
- The utility to update the bridge's firmware is only for DOS/Windows.
There will probably be LVD-SATA bridges too in the future, if SATA truly catches on.
Democracy. Whiskey. Sexy. Pick any two.
1. You can't HOTSWAP an IDE drive without risking blowing your drive, crontroller, or upsetting the powersupply.
With SATA you can.
2. You can't WARMSWAP an IDE drive, without risking blowing your drive, controller, or upsetting your powersupply.
With SATA you can.
3. IDE still only supports 2, yes 2 drivers per controller, which makes it impossible to do hardware RAID-5. That leaves us with software RAID-5 as our only option.
Who cares when you can get hardware RAID-controllers with 12 ports on one card? What is the great advantage of having the cable be the single point of failure for your whole RAID, like SCSI does?
4. IDE cables can only stretch so far, so even if you could somehow manage to get 8 IDE controllers into a box, for a total of 16 drives, there would still be cable length issues. I think 1 m is max. We need differential IDE :)
Ok, 1m can be a problem for some people. However most people do not have cases larger than 1m.
5. IDE drives are just now able to verify data integrity, but thats good since we can start using IDE drives in servers that don't need 100% uptime.
Err, why is it a problem when it is already fixed as you say?
6. ATA/100 Round IDE cables are already available. In fact I just ordered some that have a UV reflective coating for my next case mod which features a black light. Airflow isn't a big issue, in fact Compaq has been slicing up IDE cables for a long time now to increase airflow.
Round IDE-cables are expensive to produce and still large and inflexible. SATA solves it.
7. The SUSTAINED TRANSFER WRITE RATE of IDE drives is still not fast enough to store uncompressed NTSC video at 60 frames per second, or store high bandwidth Satellite streams.
So get the hardware RAID-controller and start streaming away. Oh wait, hardware RAID for SATA doesn't exist. 3ware is a figment of my imagination.
8a. Size increase (GB's) are not keeping pace with read/write access speeds and simply adding cache RAM and tweaking seek algorithms isn't going to remedy this problem.
You can't blame the interface for that. 150MB/s per drive for 12 drives on one card is way more than any SCSI solution supports -- and way more than current drives need.
8b. As, internal volatile write caches grow larger, the risk of uncommitted writes being lost in a power outage or crash increases.
So turn off the write cache. ATA supports Transaction Command Queueing although not all drives support it yet. By the time SATA drives become available, TCQ should be common.
Finally! A year of moderation! Ready for 2019?