Itanium Madison blasts Sun, IBM in encryption specs!
Bob Ceculski
Bill Todd
"Bob Ceculski" <b...@instantwhip.com> wrote in message
news:d7791aa1.03042...@posting.google.com...
> check out Madisons encryption specs ... awesome!
>
> http://news.zdnet.co.uk/story/0,,t269-s2133625,00.html
Yup, they're very good - almost as good as the new Opteron's
http://www.amd.com/us-en/Processors/ProductInformation/0,,30_118_8796_8800~6
9673,00.html ).
- bill
Bob Ceculski
who cares about opteron ... itanium runs OpenVMS, not oopsteron,
and the alpha team influences have not even hit yet ... when
they get done, oopsteron will be left in the dust ...
rob kas
You Should
. itanium runs OpenVMS, not oopsteron,
> and the alpha team influences have not even hit yet ... when
> they get done, oopsteron will be left in the dust ...
If Intel sees AMD Start making money with Opteron , You can bet
Yamill will Appear and VMS may end up on Orphan chip again.
Rob
Andrew Harrison SUNUK Consultancy
The systems vendors offering Opteron based systems are
asking x86 type pricing for them. 5K for a dual CPU
system.
HP want 20K for a single CPU 900 MHz 1.5 MB cache Itanium
server. 34K for a 2 way system.
It isn't difficult to work out paying quarter
the price for a system with 2 faster CPU's is going to
be attractive, particularly when you discover that ther
is also software available for the platform.
Regards
Andrew Harrison
Bob Ceculski
>>
> The systems vendors offering Opteron based systems are
> asking x86 type pricing for them. 5K for a dual CPU
> system.
>
> HP want 20K for a single CPU 900 MHz 1.5 MB cache Itanium
> server. 34K for a 2 way system.
>
> It isn't difficult to work out paying quarter
> the price for a system with 2 faster CPU's is going to
> be attractive, particularly when you discover that ther
> is also software available for the platform.
>
> Regards
> Andrew Harrison
yes it is paying 1/4 the price and getting junk in return ...
x86 platforms are and always will be garbage compared to
alpha ... I just hope Intel designs Itanium along the lines
of alpha and all the other terrific DEC hardware/software ...
the only thing oopsteron has going for it is the alpha
technology Palmer sold to AMD before convienently ending up
working there after gutting DEC ... as for all the other
garbage, you get what you pay for ...
Rick Jones
> The systems vendors offering Opteron based systems are asking x86
> type pricing for them. 5K for a dual CPU system.
As always, knowing what of the benchmarked config is included in that
base price is an important detail.
> HP want 20K for a single CPU 900 MHz 1.5 MB cache Itanium
> server. 34K for a 2 way system.
Got a reference for thost figures and what is/is not included in the
price? And whether that is the rx5670 or the rx2600 etc etc...
Further, HP has not (to my knowledge) announced prices for the 1.5 GHz
systems.
> It isn't difficult to work out paying quarter the price for a system
Without knowing what is/is not included in those prices you've
tossed-out, there is no way to know that the difference for the tested
configs is 1/4.
> with 2 faster CPU's is going to be attractive, particularly when you
> discover that ther is also software available for the platform.
Not that SPEC defines pricing for SPECweb99_SSL, but just how does the
280R or the Netra 20 with the 2x1.2 GHz UltraSPARC-III Cu's and that
add-on SCA 1000 Crypto card (the one where Sun is claiming up to 4300
SSL transactions per second) at a whopping 1008 SPECweb99_SSL compare?
(Yes, Sun _finally_ submitted SPECweb99_SSL figures with their SCA
1000. Nothing with the recently announced SCA 4000 though...) Looks
like with just 2GB of the 8GB used in the result, and without the
Gigaswift interfaces or the SCA, or the disc array it is at $15K.
http://store.sun.com/catalog/doc/BrowsePage.jhtml?catid=55844
Of course, pricing is always a fun game isn't it - what is in the base
config, how much to get that to the tested config, how much of the
tested config was required to get the result. For example, was the
entire 12x36GB StorEdge 3310 SCSI array and dual Ultra3 SCSI HBA
required required to hit that 1008 number, what is the pricing of ZWS
compared to Sun ONE Web Server (although it was probably the SNCA
doing all the real work as an in-kernel accelerator rather than the
web server...), how much does that SCA 1000 or the GigaSwifts cost.
All that fun stuff.
rick jones
--
firebug n, the idiot who tosses a lit cigarette out his car window
these opinions are mine, all mine; HP might not want them anyway... :)
feel free to post, OR email to raj in cup.hp.com but NOT BOTH...
Bill Todd
"Rick Jones" <f...@bar.baz.invalid> wrote in message
news:xzkra.548$bE2...@news.cpqcorp.net...
> Andrew Harrison SUNUK Consultancy <Andrew_No....@nospamn.sun.com>
wrote:
>
> > The systems vendors offering Opteron based systems are asking x86
> > type pricing for them. 5K for a dual CPU system.
>
> As always, knowing what of the benchmarked config is included in that
> base price is an important detail.
Indeed. But it doesn't change the fact that all indications are that
Opteron systems are priced *far* lower than any competition save IA32.
Processors currently cost $283 (at 1.4 GHz) to $794 (at 1.8 GHz), vs.
roughly 5x as much for the Itanic2 range ($1338 - $4200+, IIRC) offering on
average somewhat less performance (but Madison reportedly will bring Itanic
to rough equality in most areas and leadership in some without changing the
price).
>
> > HP want 20K for a single CPU 900 MHz 1.5 MB cache Itanium
> > server. 34K for a 2 way system.
>
> Got a reference for thost figures and what is/is not included in the
> price? And whether that is the rx5670 or the rx2600 etc etc...
I'm curious as well. Not all that long ago people were talking about
entry-level single-processor Itanic2 development systems from HP in the
$4500 range, with IIRC dual-processor systems starting at about $10K. But
as best I can determine prices are *far* higher than that (though I admit
that my success in navigating HP's pricing information on the Web has been
spotty at best).
>
> Further, HP has not (to my knowledge) announced prices for the 1.5 GHz
> systems.
Aha, but they *had* to when they submitted their new TPC-C score. And even
without adding up all the small items, just a basic rx5670 with 4 Madisons
plus 24 GB of memory gets the price tag above $100K.
By contrast, the Racksaver box with 32 GB of memory that nearly equals the
rx5670 McKinley TPC-C numbers costs $49K (including the small items left out
above). Its $2.76/tpmC score is the best of *any* system listed (all the
way down to inexpensive single-processor IA32 systems) and not much more
than half the $5.03/tpmC and $4.97/tpmC scores that the McKinley and Madison
rx5670 systems achieve.
SPECweb99_SSL results don't include system pricing, but HP's Madison
submission appears to be about the same configuration as the TPC-C system
(i.e., over $100K). Unfortunately, the 4-processor Opteron that reportedly
posts a slightly higher score is not yet officially listed AFAICT, but there
seems little reason to suspect that it would cost any more than the
Racksaver TPC-C system.
- bill
gregc at gregcagle.com
>
> I'm curious as well. Not all that long ago people were talking about
> entry-level single-processor Itanic2 development systems from HP in the
> $4500 range, with IIRC dual-processor systems starting at about $10K. But
> as best I can determine prices are *far* higher than that (though I admit
> that my success in navigating HP's pricing information on the Web has been
> spotty at best).
From http://www.hp.com/workstations/itanium:
zx2000 (single CPU) 512M Linux = $3298.
zx6000 (dual CPU) 512M Linux single CPU installed = $4896.
Example configs at:
http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1427
and
http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1431
--
Greg Cagle
gregc at gregcagle dot com
Andrew Harrison SUNUK Consultancy
Rick Jones wrote:
> Andrew Harrison SUNUK Consultancy <Andrew_No....@nospamn.sun.com> wrote:
>
>
>>The systems vendors offering Opteron based systems are asking x86
>>type pricing for them. 5K for a dual CPU system.
>
>
> As always, knowing what of the benchmarked config is included in that
> base price is an important detail.
>
>
>>HP want 20K for a single CPU 900 MHz 1.5 MB cache Itanium
>>server. 34K for a 2 way system.
>
The same as the HP box. 2 CPU's 4 GB of RAM 1800 GHz Opterons.
>
> Got a reference for thost figures and what is/is not included in the
> price? And whether that is the rx5670 or the rx2600 etc etc...
>
rx2600, pricing from your online pricing tool on the web.
> Further, HP has not (to my knowledge) announced prices for the 1.5 GHz
> systems.
>
1.5 MB cache not 1.5 GHZ systems.
>
>>It isn't difficult to work out paying quarter the price for a system
>
>
> Without knowing what is/is not included in those prices you've
> tossed-out, there is no way to know that the difference for the tested
> configs is 1/4.
>
>
You have it all now.
>>with 2 faster CPU's is going to be attractive, particularly when you
>>discover that ther is also software available for the platform.
>
>
> Not that SPEC defines pricing for SPECweb99_SSL, but just how does the
> 280R or the Netra 20 with the 2x1.2 GHz UltraSPARC-III Cu's and that
> add-on SCA 1000 Crypto card (the one where Sun is claiming up to 4300
> SSL transactions per second) at a whopping 1008 SPECweb99_SSL compare?
> (Yes, Sun _finally_ submitted SPECweb99_SSL figures with their SCA
> 1000. Nothing with the recently announced SCA 4000 though...) Looks
> like with just 2GB of the 8GB used in the result, and without the
> Gigaswift interfaces or the SCA, or the disc array it is at $15K.
>
Why would this be really that interesting in the general purpose
server market. SSL performance is only really a limitting factor
for web servers and only then if HTTPS is required.
And of course its a very horizontally scalable service, buying
an HP box that costs 34K with 2 CPU's hardly makes much sense
when you can buy a 5 x 2 way V210's with cryptos for the same
price.
The V210/V240 are new low cost low end 1/2U 1 GHz UltraIIIi
servers BTW, similar capacity to a Sun V280 but well under
half the price.
> http://store.sun.com/catalog/doc/BrowsePage.jhtml?catid=55844
>
> Of course, pricing is always a fun game isn't it - what is in the base
> config, how much to get that to the tested config, how much of the
> tested config was required to get the result. For example, was the
> entire 12x36GB StorEdge 3310 SCSI array and dual Ultra3 SCSI HBA
> required required to hit that 1008 number, what is the pricing of ZWS
> compared to Sun ONE Web Server (although it was probably the SNCA
> doing all the real work as an in-kernel accelerator rather than the
> web server...), how much does that SCA 1000 or the GigaSwifts cost.
> All that fun stuff.
>
What has this got to do with the pricing of the server itself. You
seem to have dissapeared off on an alarming Kerry Main type tangent
which is very unlike your normal posting style.
Regards
Andrew Harrison
Bill Gunshannon
Rick Jones <f...@bar.baz.invalid> writes:
> Andrew Harrison SUNUK Consultancy <Andrew_No....@nospamn.sun.com> wrote:
>
>> The systems vendors offering Opteron based systems are asking x86
>> type pricing for them. 5K for a dual CPU system.
>
> As always, knowing what of the benchmarked config is included in that
> base price is an important detail.
>
>> HP want 20K for a single CPU 900 MHz 1.5 MB cache Itanium
>> server. 34K for a 2 way system.
>
> Got a reference for thost figures and what is/is not included in the
> price? And whether that is the rx5670 or the rx2600 etc etc...
>
> Further, HP has not (to my knowledge) announced prices for the 1.5 GHz
> systems.
This discussion reminds me of my first (and last) HPUX system.
It was 14-15 years ago when I first came here to the University
to build a network. I needed a decent X-windows system to put
my custom managemnet tools on. I priced a number of systems and
decided the best one for the job was a SparcStation. My boss's
boss came back from a conference singing the praises of HP. He
refused to allow the purchase and scheduled an appearance by an
HP salesdroid. They brought in a system and showed how it ran
circles around the Sparc I had chosen. So my boss was forced to
spend the money we had allocated for the Sparc on a PA-RISC box
running HPUX. It was over a month late and when it arrived, it
didn't do any of the slick things that the salesdroid had been
showing off and bragging about. When we asked why the system
they sold us didn't do the stuff they had demoed we were informed
the demo system was a little bit beefier than the one we had bought.
To the tune of about %500, pricewise. I ended out with a box that
did not have the horsepower of the Sparc or even enough to do what
I needed it to do and it ended out being little more than an X-terminal
while I ran my applications on borrowed space and cpu from the Suns in
the CS department.
Moral: Be sure what you are really going to get is what you
expected and that includes pricewise.
>
> Of course, pricing is always a fun game isn't it - what is in the base
> config, how much to get that to the tested config, how much of the
> tested config was required to get the result. For example, was the
> entire 12x36GB StorEdge 3310 SCSI array and dual Ultra3 SCSI HBA
> required required to hit that 1008 number, what is the pricing of ZWS
> compared to Sun ONE Web Server (although it was probably the SNCA
> doing all the real work as an in-kernel accelerator rather than the
> web server...), how much does that SCA 1000 or the GigaSwifts cost.
> All that fun stuff.
Yes, all that fun stuff that PHB's frequently miss because they have
been blinded by a slick presentation.
bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bi...@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
David Froble
> I'm curious as well. Not all that long ago people were talking about
> entry-level single-processor Itanic2 development systems from HP in the
> $4500 range, with IIRC dual-processor systems starting at about $10K. But
> as best I can determine prices are *far* higher than that (though I admit
> that my success in navigating HP's pricing information on the Web has been
> spotty at best).
The key here may be the words "development systems". Such from DEC carried a
50% discount, and sometimes more. It could be such discounting that was refered to.
>>Further, HP has not (to my knowledge) announced prices for the 1.5 GHz
>>systems.
>>
>
> Aha, but they *had* to when they submitted their new TPC-C score. And even
> without adding up all the small items, just a basic rx5670 with 4 Madisons
> plus 24 GB of memory gets the price tag above $100K.
>
> By contrast, the Racksaver box with 32 GB of memory that nearly equals the
> rx5670 McKinley TPC-C numbers costs $49K (including the small items left out
> above). Its $2.76/tpmC score is the best of *any* system listed (all the
> way down to inexpensive single-processor IA32 systems) and not much more
> than half the $5.03/tpmC and $4.97/tpmC scores that the McKinley and Madison
> rx5670 systems achieve.
>
> SPECweb99_SSL results don't include system pricing, but HP's Madison
> submission appears to be about the same configuration as the TPC-C system
> (i.e., over $100K). Unfortunately, the 4-processor Opteron that reportedly
> posts a slightly higher score is not yet officially listed AFAICT, but there
> seems little reason to suspect that it would cost any more than the
> Racksaver TPC-C system.
People seem to forget that IA-64 has consistantly done a few things well. That
does not make them good general purpose computers.
In far past days I was known to participate in rather dubious activities. One
was motocross racing. The special purpose bikes had highly tuned exhaust
systems that for a small RPM range gave a truly stupendous HP spike. Except for
this small range, they were pretty pathetic. The trick was to keep the engine
in the small power band. These machines had one intended purpose, and were
rather poor at anything else. IA-64 and EPIC reminds me of those bikes.
Dave
--
David Froble Tel: 724-529-0450
Dave Froble Enterprises, Inc. Fax: 724-529-0596
DFE Ultralights, Inc. E-Mail: da...@tsoft-inc.com
T-Soft, Inc. 170 Grimplin Road Vanderbilt, PA 15486
Bill Todd
"gregc at gregcagle.com" <"gregc at gregcagle.com"> wrote in message
news:vas7355...@corp.supernews.com...
Hmmm. The last two citations above suggest that to get an actual running
system you need to spend a bit over $4K for a single-CPU (low-end Itanic2)
box (no monitor) and over $7K for a dual-CPU (low-end Itanic2) box (no
monitor). So I'm wondering whether the first citation refers to bare-bones
pricing (no memory, no disk, no ...?).
Nonetheless, these prices are much more like the ones I thought I
remembered. Perhaps it's just *server* pricing that's much higher (rx vs.
zx series?). Unfortunately, the Opteron systems being compared against
*are* server systems (we'll have to wait until September to see how Athlon64
systems are priced).
- bill
gregc at gregcagle.com
> "gregc at gregcagle.com" <"gregc at gregcagle.com"> wrote in message
> news:vas7355...@corp.supernews.com...
<snip>
>> From http://www.hp.com/workstations/itanium:
>>
>>zx2000 (single CPU) 512M Linux = $3298.
>>zx6000 (dual CPU) 512M Linux single CPU installed = $4896.
>>
>>Example configs at:
>>
>>http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1427
>>
>>and
>>
>>http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1431
>
>
> Hmmm. The last two citations above suggest that to get an actual running
> system you need to spend a bit over $4K for a single-CPU (low-end Itanic2)
> box (no monitor) and over $7K for a dual-CPU (low-end Itanic2) box (no
> monitor). So I'm wondering whether the first citation refers to bare-bones
> pricing (no memory, no disk, no ...?).
No, the first citation is 512M RAM, 36G disk, and low end graphics card.
If you scroll down within the pages referred to in the second citations
you can find those configurations.
> Nonetheless, these prices are much more like the ones I thought I
> remembered. Perhaps it's just *server* pricing that's much higher (rx vs.
> zx series?). Unfortunately, the Opteron systems being compared against
> *are* server systems (we'll have to wait until September to see how Athlon64
> systems are priced).
Yes - HP's "server" pricing is much higher. The zx6000 makes a more
cost-efficient computational engine in clusters if you can do without
the "server" features.
Bill Todd
> Bill Todd wrote:
> > "gregc at gregcagle.com" <"gregc at gregcagle.com"> wrote in message
> > news:vas7355...@corp.supernews.com...
>
> <snip>
>
> >> From http://www.hp.com/workstations/itanium:
> >>
> >>zx2000 (single CPU) 512M Linux = $3298.
> >>zx6000 (dual CPU) 512M Linux single CPU installed = $4896.
> >>
> >>Example configs at:
> >>
> >>http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1427
> >>
> >>and
> >>
> >>http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1431
> >
> >
> > Hmmm. The last two citations above suggest that to get an actual
running
> > system you need to spend a bit over $4K for a single-CPU (low-end
Itanic2)
> > box (no monitor) and over $7K for a dual-CPU (low-end Itanic2) box (no
> > monitor). So I'm wondering whether the first citation refers to
bare-bones
> > pricing (no memory, no disk, no ...?).
>
> No, the first citation is 512M RAM, 36G disk, and low end graphics card.
> If you scroll down within the pages referred to in the second citations
> you can find those configurations.
Mea culpa - I *did* scroll down, but in the first instance missed the
lower-priced systems and in the second missed your original qualification
that the price was for a dual-CPU-capable box populated with only a single
CPU (so technically my assertion above about the lowest dual-CPU system
price was correct, but it did not serve to contradict yours).
I should know better than to try to respond in detail when I'm rushed for
time...
- bill
Rick Jones
>> Further, HP has not (to my knowledge) announced prices for the 1.5 GHz
>> systems.
> 1.5 MB cache not 1.5 GHZ systems.
That comment was refering to the systems with 1.5 GHz Itanium 2 6M
CPUs (aka Madison), not the older McKinleys.
> And of course its a very horizontally scalable service
Indeed, web services can be quite horizontally scalable. However,
there have been no clustered SPECweb99_SSL results published to date.
That would seem to suggest that either the workload characterized by
the benchmark is either not horizontally scalable, or isn't easily
horizontally scalable. (SPECweb99_SSL does not prohibit clustered
results)
> box that costs 34K with 2 CPU's hardly makes much sense when you can
> buy a 5 x 2 way V210's with cryptos for the same price.
Do you mean to imply that it would take 5 of the two-way V210's to
match the SPECweb99_SSL performance of a single two-CPU non-SPARC
systems?
> The V210/V240 are new low cost low end 1/2U 1 GHz UltraIIIi servers
> BTW, similar capacity to a Sun V280 but well under half the price.
Sparring with you in Usenet is always fun because you are so fast on
your keyboard :) You were saying how much more expensive than the
Opterons the HP IPF systems were. I then said SPARC systems aren't
that much better compared to the Opterons (in otherwords, the pot
calling the kettle). You've come-back instead with a comparison
between the Sun systems and the IPF boxes instead of the Opterons.
So, lets then look at those V240s you brought-up (there being no
SPECweb99_SSL figures published for the V210, and I doubt you really
meant to imply that it would take 5 hardware-accelerated V210's to
match a two-CPU non-SPARC system at SPECweb99_SSL) and compare them to
those Opterons.
The SPECweb99_SSL result for a two-CPU V240 was 833 [1]. So, to match the
two-CPU Opteron results one would need two of those (actually, two and
a fraction - 1783/833 = 2.14) and to match the four-CPU result would
require 4 (actually 4.2 - ie 5).
Base price per Sun's online pricing pages for the V240 appears to be
$6,495. The SCA 500 adds another $695, so box cost - without the full
8GB of RAM - is $7190. Two of those to keep pace with the two-CPU
Opteron would be $14380, IIRC, you have asserted previously (opening
this pricing can) that the two-CPU Opteron systems (unspecified RAM)
were going for 5K? Four, dual-CPU V240s (should that really be five?)
to keep-up with the single, four-CPU Opteron system would be $28760.
Bill has taken the TPC-C pricing from the RackSaver and concluded that
the Opteron SPECweb99_SSL config would be around $49000. I do not
know if that price is truly "real" or not (seems everything is bundled
into one price, no per-line item pricing, so it is hard to check), but
for now lets go with that and then add the 8GB of RAM to those V240s
since that $49000 figure includes the RAM. That is it seems $1795 for
2, 1GB DIMMs, so 8GB of RAM would be $7180. I would trust that there
is a return credit for the four 512MB boards included in the base V240
config, so credit back that price (795x2) and so the add-on to each
system is 7180 - 1590 or $5590. So, each box is then 7190+5590 or
12,780, four of those would then be $51,120, five would be $63,900.
Heck, if we really want to have fun, then we compare the number of
Gigabit ports used by the V240-based solution - 8 to 10 - to that used
for the single-system four-CPU systems - 4. The V240's are 2U, so
four of them is 8U of rack space to the four-CPU system's 4U. Etc etc
etc...
>> http://store.sun.com/catalog/doc/BrowsePage.jhtml?catid=55844
>>
>> Of course, pricing is always a fun game isn't it - what is in the
>> base config, how much to get that to the tested config, how much of
>> the tested config was required to get the result. For example, was
>> the entire 12x36GB StorEdge 3310 SCSI array and dual Ultra3 SCSI
>> HBA required required to hit that 1008 number, what is the pricing
>> of ZWS compared to Sun ONE Web Server (although it was probably the
>> SNCA doing all the real work as an in-kernel accelerator rather
>> than the web server...), how much does that SCA 1000 or the
>> GigaSwifts cost. All that fun stuff.
> What has this got to do with the pricing of the server itself. You
> seem to have dissapeared off on an alarming Kerry Main type tangent
> which is very unlike your normal posting style.
I'm afraid I don't understand the Kerry Main reference.
If we really want to get into the ever so fun and twisty discussion of
pricing which (iric) you brought-up, and if the _solution_ put-forth
by a vendor to achieve a given SPECweb99_SSL result really did need
all those things from the SPECweb99_SSL disclosure (and there are
indeed times when a SPECweb99_SSL result had stuff it didn't really
need but just happened to be there) then comparing pricing should be
comparing _solution_ pricing not just box pricing. And after that, we
could go futher into the pricing hole and start arguing about whether
it costs more to admin four, two-CPU V240s or one, four-CPU system,
support prices etc etc etc... Probably one of the reasons why SPEC
has never been especially keen on that maze of twisty little
passages...
rick jones
[1] SPECweb99_SSL figures per http://www.spec.org/ as of April 26,
2003, SPECweb being a trademark of SPEC etc etc etc
--
Wisdom Teeth are impacted, people are affected by the effects of events.
Andrew Harrison SUNUK Consultancy
with 2 x 1 GHz 3 MB cache CPU's is 1230.
The best Sun V210/V240 number with 2 x 1 GHz CPU's is 833.
The list price of the rx2600 config used for the test was
according to your website ~46K dollars.
A Sun V240 with 8 GB and a crypto 2x 1GHz CPU's
internal disks 4 GB NICS costs 8K dollars. The
V210 would be 500 dollars less.
So at 16K for two we end up with a faster more
resiliant solution and still have 30K left in the
pot.
It doesn't take a genius to work out that you
won't be selling many at that price as web
servers and even at half the price you are
still a long way adrift.
Regards
Andrew Harrison
Andrew Harrison SUNUK Consultancy
Humm, you seem to have missed the context of the thread entirely.
We may have to compete to Opteron, allthough Sun has said that
we are interested in the CPU, not suprising since we allready
produce AMD based 32bit systems.
But the real dog fight will be between Intel and AMD in the
1-8 way x86 32 bit server space with AMD having 3 big advantages
over Intel.
1. Opteron is faster than any of the current MP x86 processors
2. Because it has onchip memory controller SMP bus and cache
like UltraSPARC IIIi AMD can charge more for it than Intel
can for the Xeons because the AMD OEMS need fewer additional
components. Or alternatively they don't have to cut so much
margin in order to respond to what ever incentives Intel offer
to keep their OEMs onside.
3. It is also 64bit while being able to run 32bit apps well.
Like it or not Intel make all their profits in the 1-8 way x86
space, IA-64 is just a huge black hole at this point.
If Opteron starts taking significant market share from Intel
in this space as many people think it will because of the 3
points above then Intel will have to respond.
They cannot respond with IA-64, it has no software its too
expensive to build in systems and it runs 32 bit apps terribly
its also slower than Opteron on Integer which dominates this
market and will probably be slower on FP once AMD publish some
x86-64bit compiler results, all of their current results
use the standard 32bit intel compilers.
Yamhill if it exists and if Intel respond with then IA-64 is
dead or close to death.
This isn't a discussion about how Sun competes with Opteron
but whether Opteron will kill your datacenter server strategy
stone dead, at the moment it isn't looking great.
> So, lets then look at those V240s you brought-up (there being no
> SPECweb99_SSL figures published for the V210, and I doubt you really
> meant to imply that it would take 5 hardware-accelerated V210's to
> match a two-CPU non-SPARC system at SPECweb99_SSL) and compare them to
> those Opterons.
>
The V210 is basically the same as the V240 but only has 1 PCI slot,
not needed for this config.
But even if this is a problem its hardly a major issue since the
V250 only costs about 500 dollars more
> The SPECweb99_SSL result for a two-CPU V240 was 833 [1]. So, to match the
> two-CPU Opteron results one would need two of those (actually, two and
> a fraction - 1783/833 = 2.14) and to match the four-CPU result would
> require 4 (actually 4.2 - ie 5).
>
As I said this really isn't a discussion about how SPARC competes
with Opteron but how IA-64 survives Opteron.
Regards
Andrew Harrison
jlsue
<Andrew_No....@nospamn.sun.com> wrote:
>
>They cannot respond with IA-64, it has no software its too
>expensive to build in systems and it runs 32 bit apps terribly
>its also slower than Opteron on Integer which dominates this
>market and will probably be slower on FP once AMD publish some
>x86-64bit compiler results, all of their current results
>use the standard 32bit intel compilers.
I make no predictions how either one will fare in the future. However, I
don't believe that either one can claim victory at this point.
To date, no AMD processor has appeared in high-end server systems for
business-class, reliable computing. This is a huge jump in mind-set for
AMD and we have no idea how well it'll do then. Benchmarks are nice, but
there's more to the story than that. If major vendors like Sun begin
building high-end servers with these processors, then we'll have something
to compare to.
Personally, I don't care about the commodity workstation market (other than
the fact that I build my own home systems, usually based on AMD CPUs). I
generally work in the mission critical computing environment and the
systems must meet some pretty stringent needs to be reliable enough to
convince me that they are viable.
To that end, this issue of 32-bit apps is, imho, a non-issue. All of the
server-based OSs that I care about are 64-bit. That's not just a niche,
it's a few billion $$ (USD) per year (HP-UX, Non-Stop Kernel, Tru64, and
OpenVMS... oh yeah, and eventually 64-bit windows).
At this point all of the systems are lower-end class systems (imho) and the
rest is all theoretical.
We can all act like talking-heads and pundits until the cows come home, but
until there's some real, live servers in full production, nobody can claim
victory or righteousness.
Andrew Harrison SUNUK Consultancy
jlsue wrote:
> On Wed, 30 Apr 2003 11:08:37 +0100, Andrew Harrison SUNUK Consultancy
> <Andrew_No....@nospamn.sun.com> wrote:
>
>
>
>>They cannot respond with IA-64, it has no software its too
>>expensive to build in systems and it runs 32 bit apps terribly
>>its also slower than Opteron on Integer which dominates this
>>market and will probably be slower on FP once AMD publish some
>>x86-64bit compiler results, all of their current results
>>use the standard 32bit intel compilers.
>
>
> I make no predictions how either one will fare in the future. However, I
> don't believe that either one can claim victory at this point.
>
> To date, no AMD processor has appeared in high-end server systems for
> business-class, reliable computing. This is a huge jump in mind-set for
> AMD and we have no idea how well it'll do then. Benchmarks are nice, but
> there's more to the story than that. If major vendors like Sun begin
> building high-end servers with these processors, then we'll have something
> to compare to.
>
It doesn't have to thats the beauty of the problem that HP
have dumped on Intel.
Intel make all their profits in the 32bit 1-8 way space.
Opteron, Althon-64, Althon-64 mobile all of which are
expected to be available this year are going to compete
directly with Intels cash cow.
Even the most optimistic estimates put IA-64 systems shipments
as 25K for this year, with slightly less than 2 CPU's per
system on average for IA-64 systems currently this puts
the total number of IA-64 processors shipped as 50K.
This puts Intel still last in the 64 bit processor space
by a very long way.
Now this would be fine if they don't loose market share to
AMD in the 1-8 way 32bit space but all the indications are
that they will.
If they lose enough than Yamhill may become reality and
IA-64 is dead.
> Personally, I don't care about the commodity workstation market (other than
> the fact that I build my own home systems, usually based on AMD CPUs). I
> generally work in the mission critical computing environment and the
> systems must meet some pretty stringent needs to be reliable enough to
> convince me that they are viable.
>
Well you should do because Intels sucess in that market competing
with Opteron will make or break their 64 bit strategy and you
work for the vendor who has tied their mid to high end systems
strategy to their current 64bit strategy.
regards
Andrew Harrison
Bob Ceculski
>
> Yamhill if it exists and if Intel respond with then IA-64 is
> dead or close to death.
>
> This isn't a discussion about how Sun competes with Opteron
> but whether Opteron will kill your datacenter server strategy
> stone dead, at the moment it isn't looking great.
>
> with Opteron but how IA-64 survives Opteron.
>
> Regards
> Andrew Harrison
and you think the 64bit chip of the high end future is
oopsteron? You cannot run a datacenter on either
oopsteron or 80,000 sparkies ... that leaves Power and
Alpha and Itanium, but if itanium gets EV8-9 fuctionality
then it will win hands down ... and remember, EV7 right
now kicks butt and it is not to late with hard work to
start EV8 back up if need be ... but if the alpha engineers
are allowed to put EV8-9 features into itanium, then itanium
would surpass EV8 because it would handle more instructions.
JF Mezei
> To date, no AMD processor has appeared in high-end server systems for
> business-class, reliable computing.
Same could be said of Intel who is also in its first venture in business
server computing.
> To that end, this issue of 32-bit apps is, imho, a non-issue. All of the
> server-based OSs that I care about are 64-bit. That's not just a niche,
> it's a few billion $$ (USD) per year (HP-UX, Non-Stop Kernel, Tru64, and
> OpenVMS... oh yeah, and eventually 64-bit windows).
That is the same mentality that killed digital: Unix will never hurt VMS
because VMS is so much better, followed by " PCs will never hurt VMS because
VMS and VAX are so much better.
The sad fact is that customers look for a box that can do the job. Why pay
500,000 for a "serious" business computing box/server when a good $2000 PC
with linux on it can run the same software ~about~ as reliably as the big
expensive machine ?
By restricting your enterprise serves to only the very high end where there
*still* exist applications that really do require those big systems, you are
doing 2 things:
-allowing the vast majority of the server market to go to other vendors,
those who sell
both low end and medium sized servers.
-what will you do when your off-the-shelf PC will have clustering etc that
gives it the capabilities at $2000 that you are charging $500,000 for on your
big boxes ?
Those who have won in the IT revolution are those who strived to lower the
cost of computing. Sun and Microsoft (for all their faults) are the big
winners (Sun in late 80s, killing DEC's workstation market) and Microsoft for
the rest.
Digital strived to protect its high end applicatiosn from being used on low
end machines to protect its cash cows. Digital no longer exists. had Digital
produced low cost workstations and gone towards higher production at lower
profit margins, Digital would still be here.
Had Digital taken the Microvax II and truly made it into a PC sized and
Apple-priced machine, DECwrite might be the "Word" of Today, and ALL-IN-1
might be the "Exchange" of Today.
JF Mezei
> oopsteron? You cannot run a datacenter on either
> oopsteron or 80,000 sparkies ... that leaves Power and
> Alpha and Itanium, but if itanium gets EV8-9 fuctionality
> then it will win hands down ...
Bob, the question isn't who will get EV8-9 finctionality. The question is who
will be able to implement and commercialise that functionality first. Concepts
are concepts.
And who will get to market first ?
IA64 with its bloated design that is different (EPIC) and delayed, slow to
market ?
Power which is a cleaner design and IBM behind it ?
8086 which is lean and mean and can be clocked to faster Mhz because of its simplicity?
Alpha had been "first to market" because the design of the architecture had
been made to be clean and easy to add all sorts of fancy widgets to the chip.
IA64 isn't a clean design, it is a compromise between Pa-Risc (and Risc chip)
and 8086 (also a risc chip) but implemented as that EPOC thing that requires
fancy compilers to get respectable benchmark results.
For as long as IA64 is playing catch up because it is delayed one generation,
IA64 won't be seen as the industry leader. Alpha may have had the lead, but
with Alpha gone, the next in line to take the leadership position is Power.
And what about Sun ? Its chip may be as slow as IA64 or even slower. But Sun
isn't affraid to market its own products and Sun isn't compromised by
Microsoft. So who will you choose ? HP with its unremarkable IA64 chip and
prducts HP won't even mention publicly for fear of competing against
Microsoft, or Sun with its unremarkable Sparc but with great marketing and no
fear of broken commitments ?
Rick Jones
> with 2 x 1 GHz 3 MB cache CPU's is 1230.
Nit - SPECweb99_SSL.
> The best Sun V210/V240 number with 2 x 1 GHz CPU's is 833.
There is no published V210 SPECweb99_SSL result. Unless and until Sun
wants to publish one it would be better to stick with the V240.
Public assertions of SPEC unpublished benchmark performance figures
for systems can result in requests for disclsoure, and since the max
RAM on the V210 is 1/2 that of the V240:
http://www.sun.com/servers/entry/v210/specs.html
It is not a given that the V210 could match the SPECweb99_SSL
performance of the V240.
> A Sun V240 with 8 GB and a crypto 2x 1GHz CPU's internal disks 4 GB
> NICS costs 8K dollars.
How did you arrive at that figure? At:
http://store.sun.com/catalog/doc/BrowsePage.jhtml?catid=100055
It lists the base price of $6,495, which includes 2GB of RAM in 512 MB
boards, the two CPUs and two, 36 GB discs. If I then go to further
configure the system (click the select box), and add 8GB of RAM (has
to use the 1 GB boards since there are only 8 memory slots), drop to
the standard warrantee (the default selection is upgrade to three year
gold for another $6200), and assume that the buyer already has OS
media and doesn't need pre-installation, that comes-out as $14,170.
Now that does not include a credit for the 2GB of RAM that needs to be
junked from the base config, assume there is one and give it that
credit ($1590) and that seems to suggest $12,580, not 8K dollars.
Where is the discrepancy? It would be good if you could include the
public URLs you are using to arrive at your pricing figures so those
folks watching can double-check our results.
> The V210 would be 500 dollars less.
And unless and until we see a V210 SPECweb99_SSL figure up on
www.spec.org we cannot assume it can achieve the same SPECweb99_SSL
result as the V240.
And even then you are still assuming an horizontal scaling for the
SPECweb99_SSL workload which has yet to be demonstrated by any
published results.
Finally, you have also left-out the cost of the web server software (a
60 day try and buy on Sun ONE doesn't cut it). Per:
http://wwws.sun.com/software/products/web_srvr/get.html
It would seem to be $1495 per CPU, or another $2990 for each V240.
Which would seem to take the price to $15,570 each.
Per:
The cost for ZWS is $1700 for an up to two CPU system.
rick jones
--
oxymoron n, commuter in a gas-guzzling luxury SUV with an American flag
Andrew Harrison SUNUK Consultancy
Rick Jones wrote:
> Andrew Harrison SUNUK Consultancy <Andrew_No....@nospamn.sun.com> wrote:
>
>>Incedentally the best HP SPECWEB_SSL benchmark result on a rx2600
>>with 2 x 1 GHz 3 MB cache CPU's is 1230.
>
>
> Nit - SPECweb99_SSL.
>
>
>>The best Sun V210/V240 number with 2 x 1 GHz CPU's is 833.
>
>
> There is no published V210 SPECweb99_SSL result. Unless and until Sun
> wants to publish one it would be better to stick with the V240.
> Public assertions of SPEC unpublished benchmark performance figures
> for systems can result in requests for disclsoure, and since the max
> RAM on the V210 is 1/2 that of the V240:
>
Ohh slap my wrist. Sorry but this forum is full of people
both inside and outside HPQ who seem prepared to take almost
any benchmark result and trail it before its announced, exrapolate
from it to something else, announce a result and then withdraw them
7 days later some of which are discouraged by the benchmark
organisations for the specific benchmark in question, some of which
are simply sharp practice.
In the circumsatances pointing out that the V210 is a less expandible
version of the V240 is a very minor misdemeanour.
> http://www.sun.com/servers/entry/v210/specs.html
>
> It is not a given that the V210 could match the SPECweb99_SSL
> performance of the V240.
>
No and its isn't a g
>
>>A Sun V240 with 8 GB and a crypto 2x 1GHz CPU's internal disks 4 GB
>>NICS costs 8K dollars.
>
>
> How did you arrive at that figure? At:
>
> http://store.sun.com/catalog/doc/BrowsePage.jhtml?catid=100055
>
> It lists the base price of $6,495, which includes 2GB of RAM in 512 MB
> boards, the two CPUs and two, 36 GB discs. If I then go to further
> configure the system (click the select box), and add 8GB of RAM (has
> to use the 1 GB boards since there are only 8 memory slots), drop to
> the standard warrantee (the default selection is upgrade to three year
> gold for another $6200), and assume that the buyer already has OS
> media and doesn't need pre-installation, that comes-out as $14,170.
> Now that does not include a credit for the 2GB of RAM that needs to be
> junked from the base config, assume there is one and give it that
> credit ($1590) and that seems to suggest $12,580, not 8K dollars.
>
Sorry forgot about the 8 GB, so the system config costs 10,680
or 21360 for 2.
The standard 4 GB system ships with 4 1GB DIMMS, we just add another
2 x 2 x 1GB DIMMS.
So we still have 25K's change from the difference between a
slower HP solution and a faster Sun one.
> Where is the discrepancy? It would be good if you could include the
> public URLs you are using to arrive at your pricing figures so those
> folks watching can double-check our results.
>
>
>>The V210 would be 500 dollars less.
>
>
> And unless and until we see a V210 SPECweb99_SSL figure up on
> www.spec.org we cannot assume it can achieve the same SPECweb99_SSL
> result as the V240.
>
> And even then you are still assuming an horizontal scaling for the
> SPECweb99_SSL workload which has yet to be demonstrated by any
> published results.
>
Hum, my suspicion is that nobody has bothered because it is a
horizontally scaled workload. And if it isn't then it isn't
terribly indicative of real world because I have a number of
customers using farms of Web servers with cryptos which do
horizontally scale.
> Finally, you have also left-out the cost of the web server software (a
> 60 day try and buy on Sun ONE doesn't cut it). Per:
>
> http://wwws.sun.com/software/products/web_srvr/get.html
>
> Per:
>
> http://www.zeus.com/purchase/
>
> The cost for ZWS is $1700 for an up to two CPU system.
>
Humm, how many customers do you have using Zeus, most will
be using Apache, SunOne or the IBM WebSphere web server.
Nice to do benchmarks with but it hardly figures in the
Web server stakes. Still it is quite cheap 1700, for
the system vs 1,495 per CPU for Sun One.
That leaves us with a total price per Server of 13670
or 27340 for two.
Sun does supply Apache supported with Solaris and the Crypto
cards also accelerate Apache as well as does the SNCA
But since you seem to want nit pick a complete costing on the
HP will need to add VxFS to the HP config price as well as the
Zeus licenses.
From your website this is $1,965 per CPU so the HP system
including software costs $51,875
Two Sun's better throughput more resiliance cost $27,340.
So we end up with 24535 change or roughly another 2 Sun's.
Lets hope you are prepared to discount the rx2600 prices
down a lot.
But you still seem to be missing the point. Opteron is your
problem because it hits your CPU partner where it hurts.
And you are almost uniquely tied to your selected CPU partner.
regards
Andrew Harrison
jlsue
wrote:
>jlsue wrote:
>> To date, no AMD processor has appeared in high-end server systems for
>> business-class, reliable computing.
>
>Same could be said of Intel who is also in its first venture in business
>server computing.
Actually, this is something with which I agree, but it also proves my point
- why do we discuss which is going to succeed at such an early point in the
lifetime of both?
Nobody can really claim right or wrong for either chip, but that hasn't
stopped the billions of bits flying around to do just that anyway.
>
>> To that end, this issue of 32-bit apps is, imho, a non-issue. All of the
>> server-based OSs that I care about are 64-bit. That's not just a niche,
>> it's a few billion $$ (USD) per year (HP-UX, Non-Stop Kernel, Tru64, and
>> OpenVMS... oh yeah, and eventually 64-bit windows).
>
>That is the same mentality that killed digital: Unix will never hurt VMS
>because VMS is so much better, followed by " PCs will never hurt VMS because
>VMS and VAX are so much better.
>
>The sad fact is that customers look for a box that can do the job. Why pay
>500,000 for a "serious" business computing box/server when a good $2000 PC
>with linux on it can run the same software ~about~ as reliably as the big
>expensive machine ?
Because it really can't. Not for mission critical apps. When you buy
those $2000 PCs, you get almost nothing for support, very little by the way
of configuration testing; and nobody will guarantee any kind of
availability.
You get what you pay for. And I don't speak this way because I'm now a
"vendor" of these products, I spend 13 years supporting them. I had
numerous problems with "good enough" crap that people would buy. Stupid,
el-cheapo wintel based pager products by
two-guys-in-a-garage-with-no-support that the departments expected 24x7
support from me on. El-cheapo gateway computers used for business-critical
SQL databases, but with no real qualification, and of course they'd buy the
cheapest memory they could find.
Businesses are considering overall IT service levels as very important for
their success. They are working more and more on IT consolidation. They
are beginning to see real value in ITIL and ITSM. El-cheapo boxes are not
going to supply the RAS features that they need to run their enterprises.
>
>By restricting your enterprise serves to only the very high end where there
>*still* exist applications that really do require those big systems, you are
>doing 2 things:
> -allowing the vast majority of the server market to go to other vendors,
>those who sell
> both low end and medium sized servers.
>
> -what will you do when your off-the-shelf PC will have clustering etc that
>gives it the capabilities at $2000 that you are charging $500,000 for on your
>big boxes ?
I don't speak for HP as a corporation, only my little slice of a career. I
do not like working on these cheap, $2K knock-offs when in a
business-critical environment. And, frankly, I'm really only interested in
the mission/business-critical environments. At this stage of the game (for
me), the other stuff isn't much fun for the amount of headaches that must
be endured to get any kind of value.
>
>Those who have won in the IT revolution are those who strived to lower the
>cost of computing. Sun and Microsoft (for all their faults) are the big
>winners (Sun in late 80s, killing DEC's workstation market) and Microsoft for
>the rest.
I don't know how you define "won"... it's not as if the "revolution" is
over by any stretch. Things are changing constantly. And, btw, many, many
of the people I've worked with who started down the "cheaper is better"
path have had to back-off and get back to the age-old "you get what you pay
for," value-based metrics.
I realize that there are still lots of enterprise level MS servers out
there. I don't dispute that. However I find that they still don't get the
level of support, and overall business value, out of those as they get with
traditional mid-range/high-end servers (and I consider Sun in that range).
>
>Digital strived to protect its high end applicatiosn from being used on low
>end machines to protect its cash cows. Digital no longer exists. had Digital
>produced low cost workstations and gone towards higher production at lower
>profit margins, Digital would still be here.
I'm not talking about protection of high-end systems. I'm talking about
providing value in enterprise-class servers, with enterprise-class
reliability, manageability, performance, and service levels.
>
>Had Digital taken the Microvax II and truly made it into a PC sized and
>Apple-priced machine, DECwrite might be the "Word" of Today, and ALL-IN-1
>might be the "Exchange" of Today.
I fully agree. Had many arguments with dec folks about the pricing on
their workstations many moons ago.
Rick Jones
> In the circumsatances pointing out that the V210 is a less
> expandible version of the V240 is a very minor misdemeanour.
I guess that depends on how much smaller a SPECweb99_SSL result it
achieves :) Still, I'm not going to lose any sleep over it.
>> Now that does not include a credit for the 2GB of RAM that needs to
>> be junked from the base config, assume there is one and give it
>> that credit ($1590) and that seems to suggest $12,580, not 8K
>> dollars.
> Sorry forgot about the 8 GB, so the system config costs 10,680
> or 21360 for 2.
So we are still apart by about 20%.
> The standard 4 GB system ships with 4 1GB DIMMS, we just add another
> 2 x 2 x 1GB DIMMS.
What is the public URL to the standard 4GB system config you are
quoting?
The Price and Buy link for the "large" V240 config from which I
started is a 2 GB config with 4x512 MB DIMMs. It listed no 4x1GB DIMM
config.
> Hum, my suspicion is that nobody has bothered because it is a
> horizontally scaled workload.
One thing I have learned (though perhaps have not always practised) is
to spell assume "ass-u-me."
I had a hand in the specification of the SPECweb99_SSL benchmark and
some familiarity with it. There is one particular aspect of the
dynamic API workload that requires a bit of work to keep the state
consistent. If you are curious, you can lookup the "post log" stuff
from the workload. There are internal consistency requirements for the
post log that require a clustered result to be more than just a bunch
of systems joined with wet string. Still there is nothing in the
workload that absolutely precludes clustering it if a vendor is so
motivated and able.
> And if it isn't then it isn't terribly indicative of real world
> because I have a number of customers using farms of Web servers with
> cryptos which do horizontally scale.
Yet someone in Sun marketing seems to think SPECweb99_SSL is at least
a worthwhile benchmark to publish, indicative or not, or we would not
have those tepid V240 and Netra20/280R SPECweb99_SSL results to
discuss here :)
The next generation of SPEC web benchmark is under development. If
you would like to contribute your thoughts, and do not already have
the contacts, I can send email to Sun's representatives to SPEC for
the benchmark suggesting they get in touch with you.
> Humm, how many customers do you have using Zeus, most will
> be using Apache, SunOne or the IBM WebSphere web server.
If the Netcraft surveys are any indication, many more folks are using
Zeus than Sun ONE. The April 2003 survey at:
http://news.netcraft.com/archives/webserver_survey.html
suggests that there many more sites running Zeus than Sun ONE.
Netcraft does not break-out an historical curve for Zeus. The current
historical curve for Sun ONE (nee Netscape) is there and shows its
considerable long-term decline.
> Nice to do benchmarks with but it hardly figures in the Web server
> stakes.
Seems to figure rather more than Sun ONE :) Both are dwarfed by Apache
and Microsoft.
> Sun does supply Apache supported with Solaris
As does HP with HP-UX. (I do not know what comes with VMS)
> and the Crypto cards also accelerate Apache as well as does the SNCA
If the SNCA is worth its salt, you aught be able to put just about any
web server behind it and see decent results since the SNCA would be
doing virtually all the work.
You would think that using crypto cards rated for thousands of
simultaneous SSL transactions per second would yield really high
SPECweb99_SSL performance. Yet that does not seem to be the case from
the results published thusfar. It shows there are still other things
that the host processor needs to do and a weak host processor cannot
hide behind a crypto accelerator the same way a web server can be
hidden behind an in-kernel accelerator.
> But since you seem to want nit pick a complete costing on the
> HP will need to add VxFS to the HP config price as well as the
> Zeus licenses.
There is a partitioning of VxFS in HP-UX. The base VxFS is included
with the OS. Certain "advanced" features are addon. None of the
advanced features appear to have been used and so the add-on Online
VxFS stuff would not be required. Also, while the OS used VxFS
filesystems (they are the default), the file_set and web server logs
happened to be on HFS filesystems. The only filesystems that see any
activity during a SPECweb99* run are those for the file set and the
web server logs.
The V240 results do not say anything about filesystem types.
Presumeably they were whatever are default for Solaris 9 4/03. The
SPEC reporting rules only require disclosure when something isn't
default.
rick jones
--
denial, anger, bargaining, depression, acceptance, rebirth...
where do you want to be today?
Bob Ceculski
now Andrew, don't go quoting spec and cert figures, as you
have stated, they mean nothing ...
as for that extra $30,000 ... that will be taken care of
by all the overtime to patch and reboot ... and that's
just for one year, so
$30000 x 4 yrs left to patch/reboot, and your solution
is $120000 on the wrong side!
Andrew Harrison SUNUK Consultancy
Now now Bob the HP boxes in question were running HP-UX
not OpenVMS so your point or at least the one you were
attempting to make with respect to OpenVMS is irrelevant.
Bit of a scratched record on your part isn't it !
Regards
Andrew Harrison