| View previous topic :: View next topic |
| Author |
Message |
Hactar
Guest
|
 Posted: Wed Sep 24, 2008 7:08 am Post subject: Re: SCSI vs SATA hard disks |
 |
|
In article <87vdwm9ssc.fsf@teufel.hartford-hwp.com>,
Haines Brown <brownh@teufel.hartford-hwp.com> wrote:
| Quote: | ebenZEROONE@verizon.net (Hactar) writes:
In article <87zllz9iip.fsf@teufel.hartford-hwp.com>,
Haines Brown <brownh@teufel.hartford-hwp.com> wrote:
An issue that has always troubled me is the optimal sequence of
partitions. For example, /swap is supposed to be located in relation to
the physical disks on the hard disk, but I never knew how to do
that. Howver, I'm not sure it makes any difference any more.
Why wouldn't it make any difference? What I wonder is: are disk
cylinder numbers arranged like
Yes, but how does one know what the physical platter limits are so that
one can put a partion near the edge?
Also, these days, I assume one does not normally use swap space,
|
I do (is that bad?), and there has been some discussion on whether
running without it is a good idea. (Conclusion: No, but not fatal.)
| Quote: | and so what effect would its location normally have? What partitions
need to be near the platter edges?
|
As you said, how do you tell what the platter edges' numbers _are_
without getting the source to the drive's firmware?
One part of my brain says swap should be near an edge to make for faster
transfers; another part says it should be toward the middle to increase
the chance that the head will pass over it when servicing other
requests. Same goes for /usr or other things for which output rate is
important.
--
-eben QebWenE01R@vTerYizUonI.nOetP royalty.mine.nu:81
PISCES: Try to avoid any Virgos or Leos with the Ebola
virus. You are the Lord of the Dance, no matter what those
idiots at work say. -- Weird Al, _Your Horoscope for Today_ |
|
| |
|
 Back to top |
Walter Mautner
Guest
|
| Posted: Wed Sep 24, 2008 7:08 am Post subject: Re: SCSI vs SATA hard disks |
|
|
Haines Brown wrote:
| Quote: | For many years I've been devoted to SCSI hard disks, but market trends
seem to suggest SCSI is dying out. If that the case, does my sticking
with SCSI any longer make sense?
SCSI is for the server market, while SATA drives mostly are consumer grade. |
SATA drives are specified for 8/24 use, while SCSI are for 24/7.
You can of course get SATA server grade drives as well, since most new
storage cabinets have moved to SATA.
The drives then are not much cheaper than SCSI drives.
| Quote: | Would a move to a SATA 3.0 Gb/s drive such as the Seagate Barracuda mean
that I will henceforth have to accept drive unreliability? I'm more
interested in reliability than top speed (or, obviously, even cost).
If I were to attach a SATA drive to my desktop machine, I'd be inclined
to do a cross install of linux to the new SATA drive from my running
SCSI drive. Any reason such a procedure would be problematic because of
the interface difference?
Both use the scsi interface layer, but you need the chipset drivers or |
drivers for your scsi/sata controller in place (kernel or initrd if you
want to boot from the attached drives).
--
vista policy violation: Microsoft optical mouse found penguin patterns
on mousepad. Partition scan in progress to remove offending
incompatible products. Reactivate MS software.
Linux 2.6.24. [LinuxCounter#295241,ICQ#4918962] |
|
| |
|
| Back to top |
david
Guest
|
| Posted: Wed Sep 24, 2008 9:56 am Post subject: Re: SCSI vs SATA hard disks |
|
|
On Tue, 23 Sep 2008 23:20:10 -0400, Hactar rearranged some electrons to
say:
| Quote: |
One part of my brain says swap should be near an edge to make for faster
transfers; another part says it should be toward the middle to increase
the chance that the head will pass over it when servicing other
requests. Same goes for /usr or other things for which output rate is
important.
|
I doubt one would be able to tell the difference. Don't sweat it.
The cylinder numbers have little to do with physical implementation on
modern hard drives anyway. |
|
| |
|
| Back to top |
Pascal Hambourg
Guest
|
| Posted: Wed Sep 24, 2008 2:30 pm Post subject: Re: SCSI vs SATA hard disks |
|
|
Walter Mautner a écrit :
| Quote: |
SCSI is for the server market, while SATA drives mostly are consumer grade.
|
What about the non-server non-consumer market, for example professionnal
workstations ?
| Quote: | SATA drives are specified for 8/24 use, while SCSI are for 24/7.
|
What does "8/24" mean ? |
|
| |
|
| Back to top |
Pascal Hambourg
Guest
|
| Posted: Wed Sep 24, 2008 3:20 pm Post subject: Re: SCSI vs SATA hard disks |
|
|
david a écrit :
| Quote: |
The cylinder numbers have little to do with physical implementation on
modern hard drives anyway.
|
However the logical address (LBA or CHS) is a rather reliable
information about how far a physical sector is from the edge of the
platters, except if the sector was reallocated because it was defective. |
|
| |
|
| Back to top |
Hactar
Guest
|
| Posted: Wed Sep 24, 2008 6:17 pm Post subject: Re: SCSI vs SATA hard disks |
|
|
In article <gbd49e$1erl$1@biggoron.nerim.net>,
Pascal Hambourg <pascal.news@plouf.fr.eu.org> wrote:
| Quote: | david a écrit :
The cylinder numbers have little to do with physical implementation on
modern hard drives anyway.
However the logical address (LBA or CHS) is a rather reliable
information about how far a physical sector is from the edge of the
platters, except if the sector was reallocated because it was defective.
|
How do you know those aren't remapped too? After all, the "C" in "CHS" means
"cylinder".
--
-eben QebWenE01R@vTerYizUonI.nOetP royalty.mine.nu:81
LIBRA: A big promotion is just around the corner for someone
much more talented than you. Laughter is the very best medicine,
remember that when your appendix bursts next week. -- Weird Al |
|
| |
|
| Back to top |
Pascal Hambourg
Guest
|
| Posted: Wed Sep 24, 2008 8:26 pm Post subject: Re: SCSI vs SATA hard disks |
|
|
Hactar a écrit :
| Quote: | Pascal Hambourg <pascal.news@plouf.fr.eu.org> wrote:
However the logical address (LBA or CHS) is a rather reliable
information about how far a physical sector is from the edge of the
platters, except if the sector was reallocated because it was defective.
How do you know those aren't remapped too?
|
What do you mean by "those" ?
| Quote: | After all, the "C" in "CHS" means "cylinder".
|
So what ? |
|
| |
|
| Back to top |
Haines Brown
Guest
|
| Posted: Wed Sep 24, 2008 8:36 pm Post subject: Re: SCSI vs SATA hard disks |
|
|
On the question of partition placement in relation to the physical
platters, I just wanted to make sure I understood.
First, I gather that these days disk geometry is translated for the
kernel in order to achieve optimization. Does this optimization mean
that the relative placement of partitions on hard disk platters is
of no longer any importance?
Second, because of this translation, I gather one can't really know the
physical placement of cylinders on the platter. So why is there a
recommendation to locate certain partitions needing fast I/O at toward
the center of the platter, when the physical location has no relation to
the start of freespace when setting up partitions?
Third, you say:
| Quote: | With a tool like /cfdisk/ you can create a partition that's aligned
with the end of the available diskspace, and thus close to the center
of the platter surface.
|
I'm unclear why the end of available diskspace would near the center of
the platter surface (I probably once knew, but have forgotten).
Fourth, if partitions requiring fast I/O should be located at the start
of the platter where cylinders have fewer sectors, does it follow they
should be located at the beginning of available disk freespace as shown
by cfdisk?
Finally, just which partitions depend on fast I/O? I understand that
with the big RAM these days, swap is not likely to be one of them. Is
the I/O issue the only factor to take into consideration when deciding
upon the sequence of partitions when using cfdisk (other than which
need to be primary and logical partitions)?
--
Haines Brown, KB1GRM |
|
| |
|
| Back to top |
Hactar
Guest
|
| Posted: Wed Sep 24, 2008 9:31 pm Post subject: Re: SCSI vs SATA hard disks |
|
|
In article <gbdm6s$1iif$1@biggoron.nerim.net>,
Pascal Hambourg <pascal.news@plouf.fr.eu.org> wrote:
| Quote: | Hactar a écrit :
Pascal Hambourg <pascal.news@plouf.fr.eu.org> wrote:
However the logical address (LBA or CHS) is a rather reliable
information about how far a physical sector is from the edge of the
platters, except if the sector was reallocated because it was defective.
How do you know those aren't remapped too?
What do you mean by "those" ?
|
LBA and CHS.
| Quote: | After all, the "C" in "CHS" means "cylinder".
So what ?
|
So whatever flaws basing decisions on cylinder number has, basing it
instead on CHS has the exact same flaws, since it's the same data.
--
-eben QebWenE01R@vTerYizUonI.nOetP http://royalty.mine.nu:81
GEMINI: Your birthday party will be ruined once again by your explosive
flatulence. Your love life will run into trouble when your fiancee hurls
a javelin through your chest. -- Weird Al, _Your Horoscope for Today_ |
|
| |
|
| Back to top |
Aragorn
Guest
|
| Posted: Thu Sep 25, 2008 4:05 am Post subject: Re: SCSI vs SATA hard disks |
|
|
On Wednesday 24 September 2008 11:30, someone identifying as *Pascal
Hambourg* wrote in /comp.os.linux.hardware:/
| Quote: | Walter Mautner a écrit :
SCSI is for the server market, while SATA drives mostly are consumer
grade.
What about the non-server non-consumer market, for example professionnal
workstations ?
|
Professional workstations typically have SAS, SCSI or servergrade SATA
disks.
| Quote: | SATA drives are specified for 8/24 use, while SCSI are for 24/7.
What does "8/24" mean ?
|
8 hours of usage per day, of which 20% under full load. SCSI is typically
rated for 24 hours a day, 7 days a week usage, of which 80% under full
load.
Do however bear in mind that if you keep your systems running 24/7, you'd be
better off with motherboards and memory modules that support ECC, because
commodity PCs don't have that and if you keep those up for too long, their
memory may get corrupted due to cosmic radiation and other EM fields,
resulting in instability.
--
*Aragorn*
(registered GNU/Linux user #223157) |
|
| |
|
| Back to top |
Aragorn
Guest
|
| Posted: Thu Sep 25, 2008 4:26 am Post subject: Re: SCSI vs SATA hard disks |
|
|
On Wednesday 24 September 2008 17:36, someone identifying as *Haines Brown*
wrote in /comp.os.linux.hardware:/
| Quote: | On the question of partition placement in relation to the physical
platters, I just wanted to make sure I understood.
First, I gather that these days disk geometry is translated for the
kernel in order to achieve optimization.
|
No, it is rather translated so that the BIOS can properly use the disk. It
has to do with the maximum number of cylinders and heads that a BIOS can
work with, and so a translation is made to make it appear as if the disk
has more heads and fewer cylinders.
| Quote: | Does this optimization mean that the relative placement of partitions on
hard disk platters is of no longer any importance?
|
Well, I'd say that the importance is to be taken with a huge spoon of salt
these days, given the CHS translations and the remapping of bad sectors by
the drive unit itself, but in overall, the closer a partition is to the
beginning of available diskspace as expressed in logical blocks, the closer
it will still be to the beginning of the hard disk.
But either way, most hard disks already have a huge data cache, and the
Linux kernel also caches and buffers a lot of data in RAM.
| Quote: | Second, because of this translation, I gather one can't really know the
physical placement of cylinders on the platter.
|
Not with accuracy, no.
| Quote: | So why is there a recommendation to locate certain partitions needing fast
I/O at toward the center of the platter, when the physical location has no
relation to the start of freespace when setting up partitions?
|
Well, as above. There is now far less accuracy in pinpointing where exactly
a partition begins and ends, but in overall partitions closer to the
beginning of a hard disk will have the highest throughput due to the outer
cylinders having more sectors per track.
| Quote: | Third, you say:
With a tool like /cfdisk/ you can create a partition that's aligned
with the end of the available diskspace, and thus close to the center
of the platter surface.
I'm unclear why the end of available diskspace would near the center of
the platter surface (I probably once knew, but have forgotten).
|
Okay, I think you misread me here.  When I say "the center of the
platter surface", I mean that the platters form a circle when seen in two
dimensions, with the spindle at their center. And with most disks that I
know of, the beginning of the storage capacity on the disk is located at
the outermost cylinders, and the end of the storage capacity is at the
innermost cylinders - leaving a few cylinders open that are even closer to
the spindle for the landing of the disk heads.
| Quote: | Fourth, if partitions requiring fast I/O should be located at the start
of the platter where cylinders have fewer sectors, does it follow they
should be located at the beginning of available disk freespace as shown
by cfdisk?
|
Yes.
| Quote: | Finally, just which partitions depend on fast I/O?
|
Well, that depends on the usage of the system. If you're running a database
server, then the partition holding the database will probably (slightly)
benefit from being closer to the beginning of the available diskspace.
But then again, this should be taken with a serious spoonful of salt because
of disk caching, and the fact that the actual type of filesystem you use
for a particular partition might have a far more significant impact on
performance than the physical location on the disk in terms of CHS
coordinates.
| Quote: | I understand that with the big RAM these days, swap is not likely to be
one of them. Is the I/O issue the only factor to take into consideration
when deciding upon the sequence of partitions when using cfdisk (other
than which need to be primary and logical partitions)?
|
What other factor would there be, beside I/O? 
And by the way, whether partitions used by GNU/Linux are primary or logical
is totally irrelevant. The primary/extended/logical thing is a DOS legacy
which may still have some relevance when setting up Windows, but it has no
value whatsoever with regard to GNU/Linux, other than with regard to what
the device special file designating a certain disk partition will be
called. GNU/Linux does not require any primary or active partitions.
I myself typically do create my first three partitions as primary, but this
is only so as to have a more consistent numbering scheme, i.e. numbers 1 to
3 for the primaries, 4 for the extended container, and 5 and upward for all
logical partitions in the extended container.
--
*Aragorn*
(registered GNU/Linux user #223157) |
|
| |
|
| Back to top |
Haines Brown
Guest
|
| Posted: Thu Sep 25, 2008 4:47 am Post subject: Re: SCSI vs SATA hard disks |
|
|
Aragorn,
Thanks for the clear explanations. I'm much more on top of the situation
now.
--
Haines Brown, KB1GRM |
|
| |
|
| Back to top |
David Lesher
Guest
|
| Posted: Thu Sep 25, 2008 5:14 pm Post subject: Re: SCSI vs SATA hard disks |
|
|
Aragorn <aragorn@chatfactory.invalid> writes:
| Quote: | I'm not so sure that's a market trend. It just so happens to be that SCSI
is no longer considered useful in the home and office desktop market, but
servers are most definitely still using SCSI.
However, the SCSI that's being used and marketed today is no longer of the
parallel variant. Just as parallel ATA had to make way for serial ATA,
SCSI has by now already started making way for serial attached SCSI (SAS)
and iSCSI for storage area networks.
|
I'm sorry, but you'll have go a long way to convince me the parentage of
SAS is anything but:
MarketDroid 1): Damn, everyone is buying SATA drives; the price is
falling and we are screwed. How do we come up with a way to charge a
premium without really doing a lot of work?
MD2: I've got it! We'll rebadge SATA into something with SCSI in the
name, so it sounds beefier... hmmm that's it.. Serial Attached SCSI.
We save the investment in ""SCSI"" and build up the hype around it.
MD1: But SATA really has some pluses.. Are we going to ignore them?
MD2: We'll use Gate's ploy -- extend and embrace! We'll tweek some
SATA specs here and there, adding some things we can talk up. But
we'll save a bundle on connectors alone.
......
In the past, SCSI server drives brought you two things: performance and
reliability. [Think of those 9 GB Barracudas..].
Now the issues are: Does SAS really do that much over SATA, for your
case? And: Does paying SAS prices really give you more reliable drives,
or just different electronics?
--
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433 |
|
| |
|
| Back to top |
Aragorn
Guest
|
| Posted: Sat Sep 27, 2008 6:08 am Post subject: Re: SCSI vs SATA hard disks |
|
|
On Thursday 25 September 2008 19:14, someone identifying as *David Lesher*
wrote in /comp.os.linux.hardware:/
| Quote: | Aragorn <aragorn@chatfactory.invalid> writes:
I'm not so sure that's a market trend. It just so happens to be that
SCSI is no longer considered useful in the home and office desktop
market, but servers are most definitely still using SCSI.
However, the SCSI that's being used and marketed today is no longer of
the parallel variant. Just as parallel ATA had to make way for serial
ATA, SCSI has by now already started making way for serial attached SCSI
(SAS) and iSCSI for storage area networks.
I'm sorry, but you'll have go a long way to convince me the parentage of
SAS is anything but:
MarketDroid 1): Damn, everyone is buying SATA drives; the price is
falling and we are screwed. How do we come up with a way to charge a
premium without really doing a lot of work?
MD2: I've got it! We'll rebadge SATA into something with SCSI in the
name, so it sounds beefier... hmmm that's it.. Serial Attached SCSI.
We save the investment in ""SCSI"" and build up the hype around it.
MD1: But SATA really has some pluses.. Are we going to ignore them?
MD2: We'll use Gate's ploy -- extend and embrace! We'll tweek some
SATA specs here and there, adding some things we can talk up. But
we'll save a bundle on connectors alone.
|
As with everything, technology is mainly developed to get marketed rather
than for progress, but SAS is far more than what you describe above.
The serialization of SCSI does offer some benefits with regard to large
enterprises and data centers, and it all falls within the spirit of
extending the possibilities of SCSI, e.g. there is also iSCSI now, which is
a SCSI tunnel over ethernet.
| Quote: | In the past, SCSI server drives brought you two things: performance and
reliability. [Think of those 9 GB Barracudas..].
|
It still does. The drives themselves - or at least, the ones I know - are
basically the same as the U320 drives, but their maximum throughput is
higher, whereas you could end up with a bottleneck on parallel SCSI chains.
| Quote: | Now the issues are: Does SAS really do that much over SATA, for your
case? And: Does paying SAS prices really give you more reliable drives,
or just different electronics?
|
SAS drives *are* SCSI drives, so they do have all the goodies that SCSI
comes with - e.g. ECC, logging, tagged command queueing - whereas SATA is
actually nothing other than a serialized ATA drive in which an attempt was
made to make ATA/IDE more SCSI-like.
Enterprise-grade SATA drives are probably just or nearly as reliable as
SAS/SCSI, but they lack the features that made SCSI stand out. SATA still
is ATA, don't forget that. Also, not all SATA drives - not even in the
enterprise-grade range - are fit to be used in RAID arrays, while SAS
drives all are RAID-rated.
On the other hand, if you care more about cost-effectiveness than features,
then SATA offers (far) more storage per Dollar/Euro than SCSI. But then
again, this was already the case for PATA - aka IDE, although SATA is IDE
as well - versus parallel SCSI.
So the bottom line is that if you're thinking about marketing scams, the
scam would rather rest with SATA than with SAS, because SATA was intended
to mimic SCSI over an IDE bus, but still has to rely on the SATA-specific
NCQ (native command queueing) over the SCSI-specific TCQ (tagged command
queueing), because TCQ on SATA sucks. Also, the difference in retail price
between a SAS disk and an U320 SCSI disk is mainly negligible.
--
*Aragorn*
(registered GNU/Linux user #223157) |
|
| |
|
| Back to top |
Brandon McCombs
Guest
|
| Posted: Mon Sep 29, 2008 1:07 am Post subject: Re: SCSI vs SATA hard disks |
|
|
Haines Brown wrote:
| Quote: | ebenZEROONE@verizon.net (Hactar) writes:
In article <87hc89bmfm.fsf@teufel.hartford-hwp.com>,
Haines Brown <brownh@teufel.hartford-hwp.com> wroteready worko), there
Would a move to a SATA 3.0 Gb/s drive such as the Seagate Barracuda mean
that I will henceforth have to accept drive unreliability?
_All_ drives are unreliable to some degree. The ultimate in
computer-readable reliability is probably Tyvek punched tape.
Yes, but my subjective impression is that there is a very wide
difference in reliability. Of the dozen SCSI drives I've used over the
years, only one failed on me; reading on line discussions and reviews,
it seems that SATA drives fail regularly.
|
I've been using SATA for the last few years and haven't had any issues
with them. I would recommend them because they provide high data
transfer rates and high RPMs with low cost. Not to mention their
connectors are so small that you can have 6 plugged into a very small
area on the motherboard and it's still easy to manage. The cables
themselves are also smaller than the older PATA cables which I also love.
I bought 2 320GB SATA2 drives 2 years ago on Newegg. I put them in a
mirror and have had no issues. My PC is on 24x7 too. About a year ago I
got 2 80GB SATA2 drives and put them into a strip array (on the same
controller as the other 2 320GB drives. Again, no issues to date. |
|
| |
|
| Back to top |
|
