SCSI vs SATA Hih-Perf

[lmanna]

Opteron is not a processor to be taken seriously ???? Any backing
with hard numbers for what you're saying ? We have a whole 64-node dual
opteron cluster running 64-bit applications for more than a year and
it's been not only reliable but given the nature of our applications
crucial in a time when Intel was sleeping in their 32-bit laurels and
convincing the industry and neophytes that 64-bit equals Itanium only.
I applaud AMD for their screw-intel approach giving floks like us a
great cost-effective 64 bit option. If the Opteron wasn't succesfull
Intel would have never come up with the 64-bit Xeon, their mantra would
have been "Buy Itanium". Have you tried to cost out a 64-node dual
Itanic lately ?? Moreover, our current file-servers are Xeon based and
we don't feel confident on their running 64-bit OS and/or XFS.

The only consideration I had for the Xeons was their wider choice of
mobo availability, and the new boards with 4x, 8x and 16x PCI-Express
options which might prevent PCI bus saturation in some extreme video
streaming or large sequential reads applications, which is not the case
in our scenario. You might also need 10GB ethernet to cope with such
data stream.

Parsifal

 

[Maxim S. Shatskih]

As somebody with now perhaps ~10 CPU years actual usage on AMD CPUs

(mostly Athlons) under Linux I cannot agree. I have had troubles, but
not a single problem because of the CPUs.

Usually, when people are speaking about "Athlons are worse", this is due to
worse qualities of _chipsets and mobos_, and not the AMD's CPUs themselves.

VIA chipsets were traditionally worse then Intel ones - for instance, in terms
of lame ACPI support.

 

[flux]

You need to catch up with the times. You are correct about the original
Itaniums being dogs, but I'm talking about the new Itanium2 processors,
which are also 64-bit. As for Intel being expensive, you get what you pay
for. The new Itanium2 sytems are SWEEEEEEET!


It's being proven in the field daily. You simple don't see Opteron based
solutions being deployed by major commercial and governmental entities.
True, there are a few *novelty* systems that use many Opteron processors,
but they are merely a curiosity than the mainstream norm. That said, if I
wanted a dirt-cheap gaming system I would opt for an Opteron based SATA box.

April Fool's a week early?

 

[Arno Wagner]

Arno Wagner wrote:

Would you do the world a favor and actually take ten minutes to research
your statements before you make them?

I maked it with a "(?)" as tentative but not sure. Still this is
a newsgroup and you get what you pay for. I also don't think "the
world" reads this group.

All SATA drives sold as "enterprise"
drives have the ability to perform staggered spinup.

It is not that easy. Depending on the mechanism, you need controller-BIOS
support or the right type of preconfiguration. Just "supports staggered
start-up" does not cut it, especially on a new product type.

Also, just to show the quality of your "research", I happen to have
found an "enterprise" disk that does not support staggered spin-up in
about 1 second: Maxtor MaxLine II plus. Staggered spin-up is only
in MaxLine III. How do I know? Because I own one of these and read the
documentation! I guess there will be more of them.

In addition I did not find any specification how the staggered spin-up
works on a MaxLine III. Does it need controller support? Is it a SATA II
only feature that does not work with an older controller? Can I jumper
it? Will the controller support be there? With SCSI I know, because it
has been a feature for decades.

Arno

 

[Arno Wagner]

Arno Wagner wrote:

Oh please, come on now! This is like saying BMW introduces a new car this
year and it is going to be a failure in the world for using cutting edge
technology that hasn't a single shred of old technology behind it. When you
lift the hood you still see the same old internal combustion engine that
they used for the last 50-years. The difference is they improved
manufacturing processes and materials to make the product better. They
didn't redesign the wheel for the sake of doing so.

Take a new Itanium2 box for a test drive and you'll open your eyes.

Oh, I agree that it is powerful hardware. But you know, I rather
have that 10 machine cluster with 10 times the storage that can actually
do the job than this single, gold-plated big iron.

Then again, if the box were being used in environments that were life
dependant such as on the battlefield, reliability is paramount over cost.
Intel has a proven track record for reliability in the field. I would feel
safe using an Intel solution over an AMD any day of the week.

So? From what I hear, getting people killed is preferred to
spending lots of money on most battlefields. And if you think
thet CPU reliability is the most important question, then
you cannot have much experience with software.

Market share has a great way of defining reliability.

Well, that is complete nonsense. Market share does not define any
technical characteristic. Market share could indicate some technical
problem, but in this instance it does not. It rather signifies
"we have allways bought Intel".

It would seem that the major players don't feel comfortable betting
their livelihood on AMD.

So? And what does that indicate exactly, besides that they just
continue to do what they always did, like any large, conservative
organisation? It does not say anything about the technological
quality of Opterons.

Define "pay too much"? Most people and I would rather pay too much
upfront instead of being backended with high maintenance and repair
costs, not to mention the disastrous outcome of total failure.

If that were so, there would be hard numbers about this out there.
Care to give a reference to a technological study that shows
that AMD is less reliable than Intel to a degree that matters?

Like I said, you get what you pay for. If the military would go
totally AMD than I would agree with you. Till that day, AMD is not
a processor to be taken seriously.

As somebody with now perhaps ~10 CPU years actual usage on AMD CPUs
(mostly Athlons) under Linux I cannot agree. I have had troubles, but
not a single problem because of the CPUs.

Arno

 

[Arno Wagner]

April Fool's a week early?

Probably suppressed machine rage. I know I have some. But then what
do I know, I use AMD CPUs and cheap drives. Probably deserve all
the problems I have ;-)

Arno

 

[Rita Ä Berkowitz]

Oh, I agree that it is powerful hardware. But you know, I rather
have that 10 machine cluster with 10 times the storage that can
actually
do the job than this single, gold-plated big iron.

Of course you would, but the majority of commercial and military entities
disagree with you.

So? From what I hear, getting people killed is preferred to
spending lots of money on most battlefields. And if you think
thet CPU reliability is the most important question, then
you cannot have much experience with software.

Sorry, software is of no concern to me since that is the other person's
problem. But, then again, there are people whom traditionally blame hardware
related problems on the software. The anti-Microsoft crowd comes to mind.

Well, that is complete nonsense. Market share does not define any
technical characteristic. Market share could indicate some technical
problem, but in this instance it does not. It rather signifies
"we have allways bought Intel".

Or more desirably, "we always sell Intel" because this is what our customers
that have a clue know what they want.

So? And what does that indicate exactly, besides that they just
continue to do what they always did, like any large, conservative
organisation? It does not say anything about the technological
quality of Opterons.

But it speaks volumes of the people purchasing the hardware. Not many want
to have egg on their face when the passing fad called the Opteron takes a
dump.

If that were so, there would be hard numbers about this out there.
Care to give a reference to a technological study that shows
that AMD is less reliable than Intel to a degree that matters?

I only go by what the majority wants and it surely isn't AMD. And most AMD
zealots wouldn't want to look at the hard numbers if they bit them in the
ass.

As somebody with now perhaps ~10 CPU years actual usage on AMD CPUs
(mostly Athlons) under Linux I cannot agree. I have had troubles, but
not a single problem because of the CPUs.

I guess it boils down to your expectations of what you want from any
particular CPU. Like I said, if it's gaming and a simple home based MP3
server for the kiddies than I'll say that AMD is the only choice from a
sheer economics standpoint.



Rita

 

[Arno Wagner]

Usually, when people are speaking about "Athlons are worse", this is due to
worse qualities of _chipsets and mobos_, and not the AMD's CPUs themselves.

In the beginning that was certainly true, especially as AMD chipsets
did not get as much R&D as the Intel ones because of low market share.
I think it is past.

VIA chipsets were traditionally worse then Intel ones - for
instance, in terms of lame ACPI support.

Agreed. I had those problems. In fact I believe it became
usable only recently. However it is not really needed on
a server.

Arno

 

[J. Clarke]

Arno Wagner wrote:

In comp.sys.ibm.pc.hardware.storage J. Clarke
Arno Wagner wrote:

In comp.sys.ibm.pc.hardware.storage (e-mail address removed) wrote:
Hello all,
[...]
One thing you can be relatively sure of is that the SCSI controller
will work well with the mainboard. Also Linux has a long history of
supporting SCSI, while SATA support is new and still being worked on.

If he's using 3ware host adapters then "SATA support" is not an
issue--that's handled by the processor on the host adapter and all that
the Linux driver does is give commands to that processor.

Do you have any evidence to present that suggests that 3ware RAID
controllers have problems with any known mainboard?

No. I was mostly thinking of SMART support, which is not there
for SATA on Linux (unless you use the old IDE driver). Normal disk
access works fine in my experience.

Actually, that would be a function of the 3ware drivers. With a 3ware
host adapter you do not use the SATA drivers, you use drivers specific to
3ware, and the 3ware drivers _do_ support SMART under Linux.

And, does that work reliably and with the usual Linux tools,
i.e. smartctl? Would kind of surprise me, since libata does
not have smart support at all at the moment, since the ATA
passthru opcodes have only very recently be defined by the
SCSI T10 committee.

Yes, it does. That is _specifically_ addressed by the 3ware drivers.

You seem hung up on the idea that the 3ware host adapters use the generic
ATA drivers. They do not, they have their own unique open source drivers.

At the moment there is no SATA command queuing under Linux, as you
can quickly discover by looking at the Serial ATA (SATA) Linux
software status report page here:

http://linux.yyz.us/sata/software-status.html

I was not saying that SATA queuing is worse. I was saying (or intended to)
that SCSI has command queuing under Linux while SATA does not currently.

This is true if you're using some 20 buck dumb-chip host adapter, not if
you're using a 3ware, which has its own built in support for command
queuing that is transparent to the OS.

[...]

Why? They see SATA as the coming thing. Are you suggesting that Western
Digital is incapable of producing a SCSI drive?

I am suggesting that WD is trying to create a market beween ATA
and SCSI by claiming to be as good as SCSI with SATA prices. If
it sounds to good to be true, it probably is.

Haven't priced Raptors have you. Pricing is a lot closer to 10,000 RPM SCSI
drives of the same capacity than to IDE drives.

It is a driver and software questtion with newer interfaces as well.
I had numerous problems with SATA under Linux.

Using an intelligent RAID controller that has its own drivers or using a 20
buck chip?

[...]

Raptors have a jumper that selects startup in full power mode or startup
in standby, intended specifically to address this issue.

Good. And does the 3ware controllers support staggered starts?
Yes.

Just tell the drive to come out of standby whenever you are ready.

That should be sometheing the controller and the drive do. Id
the OS does it, it can fail in numerous interessting ways.

So let's see, it's badness if it's not there and it's badness if it's there?
Sorry, but you can't have it both ways. The capability is there. If your
pet OS doesn't support it and you need it then get a real OS.

Of course you are free to do that. But I have 4TB or RAIDed storage
under Linux, about half of which is SATA. And I did run in the problems
I describe here.

Using Raptors and intelligent RAID controllers, or using consumer products?

You are talking about the LL drivers. There is an SCSI abstraction
layer in the kernel as well as an SATA abstraction layer. The former
is stable, proven and full-featured. The latter is pretty basic at
the moment.

So use the SCSI abstraction layer. The OS sees the RAID, it does not see
individual drives in the RAID.

To quote the maintainer:

Basic Serial ATA support

The "ATA host state machine", the core of the entire driver, is
considered production-stable.

The error handling is very simple, but at this stage that is an
advantage. Error handling code anywhere is inevitably both complex and
sorely under-tested. libata error handling is intentionally
simple. Positives: Easy to review and verify correctness. Never data
corruption. Negatives: if an error occurs, libata will simply send the
error back the block layer. There are limited retries by the block
layer, depending on the type of error, but there is never a bus reset.

We're not talking about "basic serial ATA support", we're talking about a
third-generation intelligent RAID controller that has its own drivers that
are unrelated to and do not use the SATA support built into the linux
kernel.

See above.

I saw above.

Also if specific drivers are needed for specific
hardware, they tend to be less reliable because the user-base is
smaller.

I see. So you would, given the choice between a purpose-made IBM server
with an IBM RAID controller and a consumer-crap machine made with generic
components, choose the consumer-crap because the consumer-crap drivers are
going to be "less reliable because the user-base is smaller"? Is that what
you are saying? Because you can get an intelligent SATA RAID controller
from the company that used to be IBM's RAID controller division.

I'm sorry, but the argument you are presenting would preclude the use of
_any_ intelligent RAID controller, SATA, SCSI, or otherwise.

Not in the set-up of the OP. You did read that, did you?

If he's proposing to soft-RAID using a 3ware host adapter then he's a damned
fool.

Seems to me we have a misunderstanding here. If the OP
wanted to do Hardware-RAID the assessment would look
different.

He's the one who stated that he was going to use the 3ware host adapter. If
he doesn't want to make use of its abilities then he should save his money.

 

[J. Clarke]

In comp.sys.ibm.pc.hardware.storage J. Clarke



I maked it with a "(?)" as tentative but not sure. Still this is
a newsgroup and you get what you pay for. I also don't think "the
world" reads this group.


It is not that easy. Depending on the mechanism, you need controller-BIOS
support or the right type of preconfiguration.

The same is true of SCSI. So what?

Just "supports staggered
start-up" does not cut it, especially on a new product type.

Also, just to show the quality of your "research", I happen to have
found an "enterprise" disk that does not support staggered spin-up in
about 1 second: Maxtor MaxLine II plus. Staggered spin-up is only
in MaxLine III. How do I know? Because I own one of these and read the
documentation! I guess there will be more of them.

Take a look at the back of the drive. You should see a jumper block next to
the SATA connectors. Putting a jumper on the right pins in that block
enables staggered spin up. It appears that there is documentation that you
did _not_ read.

In addition I did not find any specification how the staggered spin-up
works on a MaxLine III. Does it need controller support? Is it a SATA II
only feature that does not work with an older controller? Can I jumper
it? Will the controller support be there? With SCSI I know, because it
has been a feature for decades.

Look harder. It's on the Maxtor site. It took me about 30 seconds to find
it, most of which was waiting for pages to load.

 

[Curious George]

I wouldn't use either one of them since your major flaw would be using an
Opteron when you should only be using Xeon or Itanium2 processors. Now, if
you are just putting an MP3 server in the basement of your home for
light-duty work you can squeak by with the Opterons. As for the drives, I
would only use SCSI in the system you mention.

Rita,

You've got to be the most predictable poster on usenet. Many of us
would choke if you ever made different points, or sold Intel & scsi
based machines without trolling.

 

[Rita Ä Berkowitz]

You've got to be the most predictable poster on usenet. Many of us
would choke if you ever made different points, or sold Intel & scsi
based machines without trolling.

LOL! Is there really anything else worth using besides an Intel based SCSI
system? Point made!



Rita

 

[Folkert Rienstra]

Not Hitachi, not Seagate, not Maxtor, that are SATAII compatible.

The PSU needs to deal with that anyway when all drives are seeking randomly.
Seagates appear to spinup agressively though.

Would you do the world a favor

Babblemouth? Never!

and actually take ten minutes to research
your statements before you make them?

He'll die first.

All SATA drives sold as "enterprise"
drives have the ability to perform staggered spinup.

So do all Hitachis and Seagate Barras 7 and 8 and Maxtor Diamondmax+9, +10

 

[Arno Wagner]

Arno Wagner wrote:

Take a look at the back of the drive. You should see a jumper block next to
the SATA connectors. Putting a jumper on the right pins in that block
enables staggered spin up. It appears that there is documentation that you
did _not_ read.

Well, according to the product manual the jumpers on the SATA version
have no functionality. I am not about to set unmarked jumpers. Also
these drives, while sold as "enterprise drives" are not SATA II drives.
Maybe that is what confuses you. Staggered spin-up is optional in SATA I
drives.

Anyways, I think there is no point in continuing this.

Arno

 

[J. Clarke]

Well, according to the product manual the jumpers on the SATA version
have no functionality. I am not about to set unmarked jumpers. Also
these drives, while sold as "enterprise drives" are not SATA II drives.
Maybe that is what confuses you. Staggered spin-up is optional in SATA I
drives.

What confuses me is that you are saying something different from what Maxtor
is saying. Go to the Maxtor site and plug "staggered" into their search
box. I had hoped that I had given you enough of a hint for you to figure
out on your own that maybe you should do this before digging yourself in
deeper.

Anyways, I think there is no point in continuing this.

Then don't.

 

[Arno Wagner]

Arno Wagner wrote:

What confuses me is that you are saying something different from what Maxtor
is saying.

O.K., I think you are talking about the white-paper on staggered spin-up.

The problem with that is that the white paper is (1) the only place
where this behaviour is mentioned (2) the first table column of Table
3 is inconsistent. Either the pins are for storage of unused jumpers
or for delayed spin up. Also there are 8 possibilities for "installing
a jumper" on these seven pins. Figure 2 in the same white-paper gives
an entirely incorrect drawing of the connector on the drives. This
drawinf may be correct for MaxLine III or DiamondMax 10, but it is not
for MaxLine II and DiamondMax 9.

Sorry, but I think that this white paper just gives wrong information.

Arno

 

[Eric Gisin]

It turns out Arnie knows little even when the topic is Linux.

There is a Linux SATA FAQ: http://www.linuxmafia.com/faq/Hardware/sata.html

There is an interesting claim regarding RAID - none of the SATA1 chips can
support hotplug. Is this true?

I see there is an AHCI SATA2 driver, dunno what state it is in. This should
support the nForce4 as well as ICH6/7. With port multipliers, this would be
the best way to do software RAID (no extra cost), with bandwidth of 6-12Gb/s
on four SATA2 ports.

Note that Longhorn also has AHCI drivers, which may be backported to Win 2K3.

For a review of SATA RAID cards, see http://www.tweakers.net/reviews/557/1 .

 

[Curious George]

LOL! Is there really anything else worth using besides an Intel based SCSI
system? Point made!

For x86 "Servers" & "Workstations" I'm also a Supermicro slut and a
Seagate SCSI bigot. These are safe bets for a stable, reliable
platform; their quality & consistency makes integration easy & yields
good value. Believe it or not, though, there are other worthwhile
things & an inflexible one size fits all approach is inherently
flawed. But that's not my point. Your answer proved it nonetheless.

 

[J. Clarke]

Arno Wagner wrote:

Arno Wagner wrote:

In comp.sys.ibm.pc.hardware.storage J. Clarke
[...]
Just "supports staggered
start-up" does not cut it, especially on a new product type.

Also, just to show the quality of your "research", I happen to have
found an "enterprise" disk that does not support staggered spin-up in
about 1 second: Maxtor MaxLine II plus. Staggered spin-up is only
in MaxLine III. How do I know? Because I own one of these and read the
documentation! I guess there will be more of them.

Take a look at the back of the drive. You should see a jumper block
next to
the SATA connectors. Putting a jumper on the right pins in that block
enables staggered spin up. It appears that there is documentation that
you did _not_ read.

Well, according to the product manual the jumpers on the SATA version
have no functionality. I am not about to set unmarked jumpers. Also
these drives, while sold as "enterprise drives" are not SATA II drives.
Maybe that is what confuses you. Staggered spin-up is optional in SATA I
drives.

What confuses me is that you are saying something different from what
Maxtor is saying.

O.K., I think you are talking about the white-paper on staggered spin-up.

The problem with that is that the white paper is (1) the only place
where this behaviour is mentioned (2) the first table column of Table
3 is inconsistent. Either the pins are for storage of unused jumpers
or for delayed spin up.

Perhaps it has not occurred to you that among the 9, not 7, jumpers there is
one position that is "dead" and provided for jumper storage, a not uncommon
configuration, and another that enables staggered spinup. Why would they
provide a rack for unused jumpers if the drive does not have any use for
jumpers?

Also there are 8 possibilities for "installing
a jumper" on these seven pins.

Actually, I believe there are 12.

Figure 2 in the same white-paper gives
an entirely incorrect drawing of the connector on the drives. This
drawinf may be correct for MaxLine III or DiamondMax 10, but it is not
for MaxLine II and DiamondMax 9.

Sorry, but I think that this white paper just gives wrong information.

When you know for sure get back to us. You might also want to tell Maxtor
about it.

But at this point even if you are correct about that specific drive you're
arguing minutiae.

Personally if I had a Maxline II Plus that I wasn't using I'd try the damned
jumpers and see if any position gave the desired effect.

 

[Rita Ä Berkowitz]

For x86 "Servers" & "Workstations" I'm also a Supermicro slut and a
Seagate SCSI bigot. These are safe bets for a stable, reliable
platform; their quality & consistency makes integration easy & yields
good value. Believe it or not, though, there are other worthwhile
things & an inflexible one size fits all approach is inherently
flawed. But that's not my point. Your answer proved it nonetheless.

I see you learned my tastes? Yes, I realize there are other "worthwhile"
solutions out there. That's not the issue since I put all options on the
table for the customer. If I don't have something to fit their needs I
refer them to people that do. I see no logic in pissing around with
hardware that has no benefit for my customers or myself.



Rita