I am using consumer-grade SATA drives on a consumer-grade
motherboard/controller, and they work perfectly fine. So all I can
add is that SATA is not completely unreliable.

That quote doesn't give any real information, so there's no way to
evaluate its truth. Why don't you ask if anyone is currently using a
system with SATA like you intend to build, and then you'll find out if
they had any problems with the SATA part.

There are 2 segments, desktop & enterprise:
o Enterprise drive engineering is focused on Reliability + Performance
---- SCSI is chosen for bus bandwidth & multi-drive capability & reliability
o Desktop drives engineering is focused on Cost + Capacity + Appropriate-Reliability
---- ATA/SATA is chosen for chipset cost, cabling cost, appropriate

SATA is a mess by virtue of it being trying to be all things to all people.
o SATA plans on integrating SCSI techniques (TCQ) outside of SAS
---- that's a 20-50% benefit on multi-small-random-access (MS-IE to icon files)
o SATA drives may not however be engineered like SCSI 24/7/365 thrashing
---- so this is a desktop benefit from SCSI-enterprise-filters-down-to-desktop

That said the mkt is moving to smarter use of cheap h/w for certain segments.
o SATA drives are very low cost - but lesser reliability than SCSI
o So combine multiple low-cost drives with 3ware-RAID to get higher reliability

Hence d2d backup servers, NAS, etc using SATA drives.
o Enterprise - if a server goes down, image recovery faster by d2d than tape
o Consumer - WirelessAP differentiate by remote auto-backup file server

The interface is one thing, the mechanical spec of the drive quite another.

The plan seems to be:
o SATA for desktop, SAS for enterprise
---- thus far most 3.5" SATA drives use a bridge to ATA chipset
---- the 2.5" SATA Fujitsu drive I think is the first not to - and adds TCQ
o SATA & SAS use same data cabling & connectors
---- SAS can allow dual porting to the drive for redundancy
o SATA & SAS plan on the same host adapter even, just protocol difference
---- so enterprise with SAS HBA can mix SATA in if required

Only a committee of taxi drivers could come to this objective by this method.

I'd have just created SAS, and dumped SATA.
o Desktop has 1 interface (SAS), but 2 drive standards (Desktop or Enterprise)
o So drive standards defines mechancial reliability/performance
o So drive standard also defines feature-set re TCQ

Instead we have a lot of half-way house offerings.
In the meantime we do have cheap SATA drives for 3ware RAID 12-port boxes,
themselves redundantly arranged to create a good high-speed tape substitute in
terms of fast recovery of data, with tape used for library offlining re access time.

I'm still wondering how the connectors/cabling will scale - and myself waiting to
see how SATA is implemented in 2.5" re side-or-rear-mounted connectors.
Can't the SAS development team just invade the SATA team & regime change?
--
Dorothy Bradbury
http://homepage.ntlworld.com/dorothy...ry/panaflo.htm (Direct)

"Dorothy Bradbury" <dorothy.bradbury@ntlworld.com> wrote in message
news:IXVbc.598$V13.89@newsfe1-win...
Cite any source that suggests that actual drive design....heads+actuators,
platters and spindle bearings are technically any different betwen the top
SCSI 'enterprise' drives and SATA drives like the Raptor. You can't because
the technology at any given drive generation is the same and is moving fast.
There aren't TWO fundamental designs.


That's interface and controller card stuff and has nothing to do with drive
physics.


Just no.


HUH, only if there are a number of outstanding IOs, which gives a sluggish
workstation but possibly a high performance transaction server. SATA
command overhead is lower giving it an advantage on workstations operated
below IO saturation.


The above however is based on wild and false conjecture.

There's no indication that good SATA HDs are of lower reliability than SCSI
HDs. Note that the warranty length is NOT an indication of reliability.


Now you're gettin it.


Exactly.



Well Seagate, Intel and SiliconImage demo-ed it first.

Seagate claim TWO fundamental drive design segments:
o Proceedings of 2nd Annual Conference on File & Storage Technology (FAST)
o March 2003
o Seagate Whitepaper
http://www.seagate.com/content/docs/...CSI_042003.pdf

It details a host of engineering differences between desktop & enterprise class drives.

o TWO fundamental designs - Enterprise & Desktop
o Enterprise drives are mechanically different to Desktop drives
---- Enterprise *drives* target higher reliability & performance
---- Desktop *drives* target higher capacity & cost competitive
o Drive application segment goes *beyond* interface
---- Raptor = Enterprise segment, Cheetah = Enterprise segment
---- former uses SATA, latter SCSI - *both* are Enterprise segment


o What's a "good" SATA? A Raptor 10,000rpm? Well that's an Enterprise drive.
o What's a SCSI HD? A Cheetah 10,000rpm? Well that's an Enterprise drive.

Distinction is the *drive design* segment - enterprise or desktop.

Indeed drive design segment is spreading into the 2.5" market:
o Hitachi now produce an enterprise class & laptop class 2.5" 7200rpm HD
o Enterprise class (EK) version = continual use rated, laptop class = is not


No, it's based on two points:
1) Two drive design segments exist
---- Desktop & Enterprise
2) Current *market offerings* bias the former - at the moment
---- most SATA drives are desktop drives - irrespective of the interface
---- some SATA drives are *enterprise drives* - irrespective of the interface
-------- a Raptor is a *drive* designed for enterprise use
---- most SCSI drives are enterprise drives - irrespective of the interface

Seagate's point - and WD with Raptor - is that Drive-Design-Differs:
o Yes the Raptor has an SATA *interface*
o However the Raptor is an *Enterprise drive* in terms of design

A potential problem is in the implementation of a SATA system:
o SATA Desktop solutions exist - Highpoint
o SATA Enterprise solutions exist - 3ware

SATA or SCSI alone doesn't mean delineate Enterprise or Desktop.
That ignores the cost:benefit of Raptor + 3ware = cheap multi-TB.
o Raptor doesn't win just because of Cheap + Multi-GB + SATA
o Raptor wins because it is Enterprise-Class *as well* = Substitution

SATA drives may not be engineered like SCSI - you have to compare
underlying drive technology re apples to apples, enterprise to enterprise.

o The interface doesn't determine the class of drive
o The design of the drive determines its class

Ok, some will still argue SCSI is a superior enterprise interface to SATA.
That is likely to be a depreciating argument - as SATA & mkt offerings change.

I do think SATA is a mess - but mainly from the low-end implementation:
o Desktop ATA drives using a SATA bridge chip
o Desktop ATA controllers which are just that - desktop use
o SATA connectors aren't well latching
o SATA should have launched with multi-drives per channel

SATA *was* urgently needed - as anyone who has implemented 18" ATA
cable length limits with an 8-port 3ware card re routing & drive-bay distance.
Enterprise can come in a SATA interface - Raptor & 3ware prove it.

I hope that's clearer - there *is* a desktop v enterprise drive design difference.
Most SATA drives are *desktop drives*, Raptor is an *enterprise drive*.

Ok, perhaps Seagate are lying their ass off and we've been overpaying thro
the nose for years for Enterprise class drives which were the same as Desktop.
Perhaps, however enterprise drives seem to outlast desktop in the same task.
Therein is the marketing & engineering win for Raptor over other SATA drives.
--
Dorothy Bradbury

Important to distinguish between consumer & enterprise.

SATA seems to be getting slated as much re consumer solutions:
o Consumer -- Highpoint SATA RAID & cheap SATA desktop drive
o Enterprise -- 3ware SATA RAID & Raptor SATA enterprise drive

The gap isn't so much SATA, as the solutions chosen within that interface.

Considering a lot of consumers want 2-port data RAID-1 or RAID-0,
I still wonder why they don't pay the bit extra for 3ware over Highpoint.
--
Dorothy Bradbury

"Dorothy Bradbury" <dorothy.bradbury@ntlworld.com> wrote in message
news:%mAcc.10865$4N3.6887@newsfe1-win...

First of all let's look at the title of the paper:
"More than an interface - SCSI vs. ATA"

That belies the distinction you suggest below between Enterprise and
desktop/consumer. This paper is about ATA vs SCSI. SCSI is a big markup
item for the mfgs and they try to protect their profit margins with pure
hype. The first sentence noted below simply confirms my assertion. My
information comes from within a drive design group but not Seagate.

"This paper sets out to clear up a misconception prominent in
the storage community today, that SCSI disc drives and IDE
(ATA) disc drives are the same technology internally, and
differ only in their external interface and in their suggested
retail price. The two classes of drives represent two different
product lines aimed at two different markets."

"...the manufacturing and testing process..."

There may in fact be some differences here but wont affect drive reliability
in the far field..

"A Smart city coupe from DaimlerChrysler is much different than a Mercedes
E-class
sedan,..."

The clincher. This paper has nothing to do with engineering and science but
is a marketing paper built on this kind of crap.

"The most important quality in PS drives is that a drive have a
cost commensurate with the cost of the system in which it is
installed. The cost pressure of the personal computer market
gave rise to the first low-cost hard discs, and has continued
to put pressure on PS drive pricing. As we discuss PS drives,
we will come back to this point repeatedly: low cost dominates
the design of PS drives."

This point is quite correct. It's all about cost and keeping up with the
competition and sales volume. It's a RPM, platter density, noise, head
data channel data rate and heat race. Reliability is a fundamental
prerequirement. All this goes into why that are NOT two fundamental designs
at a given RPM and drive generation.

"ES drives tend to
drive costly innovation - achieving new levels of performance,
reliability or function - and PS drives adopt that technology
when it becomes cheap enough."

True. There are 15K RPM SCSI enterprise drives. There are not 15K RPM SATA
drives. HOWEVER the issue is same generation SCSI 10K RPM vs SATA 10K RPM.
There the differences disappear primarily.

"When one drive is trying to seek or
simply stay on track while nearby drives are spinning, there
is an energy transfer, known as rotational vibration, from
one seeking drive to the other drives in the cabinet."

A red herring issue for recent generation drives. This is something from 5
years ago and before and not currently relevant.

So enough. Anyone can go and read the paper carefully and see that it's
basically marketing crap.


Just names based not on technology nor science but only marketing and price
point.



Right, they simply name it differently and give the enterprise version a
longer warranty and charge bigger bucks...that's mostly it.


Nope.


Nope, only in terms of name. HOWEVER WDC has chosen to both describe the
drive as "enterprise" AND price it at comsumer/desktop levels. That sends
shivers down the back of the SCSI HD market.

Huh?

Now, you're beginning to get it.

Now, you're beginning to get it.

And it IS for some configurations.

Rethink what you're saying. Do you work for a HD mfg?


Nope, they weren't the same. They were SCSI and they were offered with a
longer warranty and they were produced in lower volume. The hype occurred
when the SCSI HD zealots oversold the advantages of SCSI and projected them
into small/modest server and highend workstation environments. There SCSI
often is NOT the best price performance solution. That became apparent over
three years ago. Some of the old wives' tales continue to hold on.

"Dorothy Bradbury" <dorothy.bradbury@ntlworld.com> wrote in message
news:g8Bcc.10870$4N3.8213@newsfe1-win...
Yes but do so accurately. See my other post.

Why does the above sound convoluted and like double talk?

Huh? Check the prices. There are very good solutions from Promise and
Intel ICH5R and SiliconImage.

"Dorothy Bradbury" <dorothy.bradbury@ntlworld.com> wrote in message
news:%mAcc.10865$4N3.6887@newsfe1-win...
Dorothy, thanks for the great information. Unfortunately, Rodney still
thinks that there is onboard termination on LVD drives. But, I do agree
that SATA is still a tad behind SCSI at this point. Like anything else, it
will get better over the years.

Rita

The SCSI "zealots" must be onto something. You don't see many, if any,
posts in here that their 10K or 15K Cheetah died. Seagate's 5-year warranty
is also great. How come you don't see this type of protection with ATA and
SATA drives? With the inexpensive price of U160 and U320 combined with a
5-year warranty it's a hell of a better deal going SCSI. Generally, before
you get to use this warranty the system the drive is in is obsolete. I must
ask. Why would anyone want to make such large compromises and jeopardize
their data, time, and productivity using ATA and SATA?



Rita

I do too - I am really looking forward to 2 things:
o 2.5" SAS -- compact cable + cheaper cable/termination + fast
---- Seagate Savvio - 10,000rpm albeit at a higher price
o 2.5" SATA -- may delivery it all, quicker, cheaper & sooner
---- Hitachi 7200rpm - here in ATA + Extended-use version + rugged

I will end up using 2.5"-SATA re multiple vendors & G rating.

SATA connector annoys re size & side-positioning - ATA allowed the
short-side-vertical (70mm) mounting, SATA forces long-side-vertical (100mm).

SATA-2 will really begin to get the ball rolling re daisy-chaining a la SCSI
of cheap inexpensive drives - without having to pay for 8-port $$$s cards.
--
Dorothy Bradbury

"Dorothy Bradbury" <dorothy.bradbury@ntlworld.com> wrote in message
news:UvFcc.10973$4N3.3631@newsfe1-win...

The Seagate 73GB drives look nice.


This is why nothing beats an SCA or FC backplane.


Yeah, I never did see the logic in building a squirrel's nest of wires to do
get half the yield of a single SCSI cable. Plus, paying extra to have an
unreliable system defies all logic.



Rita

"Dorothy Bradbury" <dorothy.bradbury@ntlworld.com> wrote in message news:UvFcc.10973$4N3.3631@newsfe1-win
Oh? Why is that?

There is no daisy chaining with SATA (nor SAS).
You will need a port multiplier or some other "concentrator" (SATA) or an "expander" (SAS).

Unless they catch on very well, they will be anything but cheap.

"Dorothy Bradbury" <dorothy.bradbury@ntlworld.com> wrote in message news:IXVbc.598$V13.89@newsfe1-win
So was ATA already.

Whatever that is supposed to mean.

SCSI drives as well may not be engineered like SCSI 24/7/365 thrashing.

Almost (for internal connections). There is a nodge in the SAS connector
to prevent it from being used with a SATA connector. Apparently it is
possible to connect a SATA to a SAS (internal, SATA style)connector.

Redundancy of what?

Yes, but requires extra HW, but yes, SAS allows for connecting SATA
drives through bridges with help of the Serial ATA Tunneled Protocol (STP).

o Fujitsu 2.5" SATA drives have a side mounted connector
---- vertical drive & rear cable exit = drive long-side is vertical (100mm)
o All other 2.5" ATA drives have a rear mounted connector
---- vertical drive & rear cable exit = drive short-side is vertical (70mm)

The Fujitsu 2.5" SATA could be mounted short-side vertical, but that will put
the SATA cables on the top/bottom taking up space there even if rt-angle plugs.

o It could be Fujitsu are alone in side-mounting the SATA connector.
o I've not checked 2.5" drive dimensions re rear-mounting SATA data/power

I have seen 2.5" SATA RAID backplanes in a single 5.25" form-factor:
o Drives were positioned side-by-side & backplane on the rear
o That either implies rear-mounted connector for the backplane
o Or, a thin-film connector from the side to the rear for the backplane

Other 2.5" SATA RAID backplanes mount the drives pushed sideways
into the enclosure - 2.5" 100mm-length to match 3.5"-drive 100mm width.
--
Dorothy Bradbury

Trying to get the physical layer of SAS, at t10...
o www.t10.org/ftp/t10/drafts/sas1/sas1r04.pdf

So far proving reluctant to download.

I thought the LT objective of both SATA-2 & SAS was the ability to
physically-chain-drives like present day (parallel) SCSI or (parallel) ATA?
--
Dorothy Bradbury

Finally got it at t10.
o Yes indeed - it does require Expander devices for SAS
o p76 gives a decent picture at the URL I gave (p108 of 495)

I notice the following covers 2.5" form-factor...
o SFF-8223, 2.5" Drive Form Factor with Serial Connector

Gone to hunt that down - Fujitsu 2.5" SATA are side mounted,
and SFF-8323 will also be interesting.

Yes, SAS internal cables will use SAS cable receptable on the SAS
target device, but SATA-style cable receptacle on initiation/expander.

Thanks for that.
--
Dorothy Bradbury

Older ref of SFF-8223 ... http://www.t10.org/ftp/t10/document.02/02-166r0.pdf
Need to check a later document, that is Apr'02, however...

o 2.5" drive form factor design schematic on p12/13 shows
---- 2.5" connector width (A2) within the housing width (A1)
---- housing width (A1) is defined as 69.85mm
---- thus the connector is obviously on the short-side (end) of a 2.5" HD

o Matches the designs of 5.25" backplane/RAID-cage for 2.5" I've seen
---- several layers of twin 2.5" drives side by side
---- using a short-side mounted connector to the backplane

Need to check on connector dimensions & bend radius next tho.

Thanks Rienstra for challenging.
--
Dorothy Bradbury

Well, I am using the WD360GD "Raptor" enterprise drives for a desktop
machine. I have them in RAID-0, 36GB + 36GB = 72GB. This drive model
does not support command queueing. As I understand, command queueing
optimizes the seek order for a batch of disk operations, which
improves performance for large batches of random, small-sector
reads/writes typical in a server. The WD740GD 74GB is the first
"Raptor" SATA model to support command queueing. SCSI has had command
queueing for a long time.

http://www.hardwarezone.com/articles...cid=10&id=1018

These drives are both SATA, and they are reliable enough to sport a
five year warranty. My point is that to say "SATA is generally
unreliable" is uninformative and misleading, unless you back it up
with some facts. And once you back it up with facts, means you narrow
it down to a particular drive model, controller, usage pattern, etc,
and it is not a general statement anymore. So far in this thread, I
have seen very few facts.

In that regard, I did have a SATA Hitachi Deskstar die on me recently,
within a week of installing. A second one gave weird SMART readings,
but did not die. These drives had a three-year warranty, while I've
seen some other SATA drives with only a one-year warranty.

BTW, people talk about Desktars being so fast, but my 7K250 was not
noticeably faster than a Western Digital 120 GB WD1200JB ATA-100
drive, and didn't bench much faster either. The Raptors on the other
hand, are noticeably faster, and in RAID-0, they fly.

When I phoned Fujitsu in early 2003, they said the SATA connector
was no longer on the end of the drive - but on the side of the drive.
On querying further they said side meaning long-side, not short-end.

Of course, SATA is incorporated into the side of the drive by virtue
of the way the power & data connectors are physically fitted into it.

Email'd photo of a Fujitsu SATA drive indeed shows it on the "end".

Well I laughed - better than crying I suppose :-)
--
Dorothy Bradbury

Random, small-sector reads, is also perhaps a good description
of MS-IE small files, icon files & so windows desktops.

Yes, the Raptor shows you can get the cake - and eat it.
--
Dorothy Bradbury