Re: Itanium Madison blasts Sun, IBM in encryption specs!

From: Rick Jones (foo_at_bar.baz.invalid)
Date: 04/30/03 

Date: Wed, 30 Apr 2003 18:59:47 GMT

Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_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Public assertions of SPEC unpublished benchmark performance figures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:

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:

http://www.zeus.com/purchase/

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
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...

Relevant Pages

  • Re: Why Apple RAM is expensive
    ... Crucial (the best non-Apple RAM). ... DRAM capacities are growing exponentially, and test times will grow in proportion to the size of the DRAM unless steps are taken to reduce test time. ... If test times grow too large, testing could become a limiting cost in the production of a DRAM. ... The question will remain if either crucial or Kingston can afford high priced test equipment. ...
    (comp.sys.mac.advocacy)
  • Re: Why Apple RAM is expensive
    ... Crucial (the best non-Apple RAM). ... DRAM capacities are growing exponentially, and test times will grow in proportion to the size of the DRAM unless steps are taken to reduce test time. ... If test times grow too large, testing could become a limiting cost in the production of a DRAM. ...
    (comp.sys.mac.advocacy)
  • Re: Why Apple RAM is expensive
    ... Crucial (the best non-Apple RAM). ... There seems to be a general consensus that testing can be a significant portion of the cost of a DRAM. ... DRAM capacities are growing exponentially, and test times will grow in proportion to the size of the DRAM unless steps are taken to reduce test time. ...
    (comp.sys.mac.advocacy)
  • Re: X360 today. My thoughts....
    ... My CPU cost me £75 at the time, and I re-used a lot of items from the last PC, including case, drives etc. ... My machine is nearly 2.5 years old and, yes I have added a larger HD, and doubled the RAM to 1GB. ... Now I know there's some sort of bias to nVidia cards so on two near-specced cards the ATi will fare worse, but is the difference really that big? ...
    (uk.games.video.xbox)
  • Re: disk or RAM
    ... run a benchmark to prove the mettle of his machine, ... an "old timer" might tell you the buffer in RAM is simply a multiple of the record blocking of the data file which is largely determined by the track/sector mapping of the particular disk. ... If the file was read sequentially it was necessary to provide enough buffering that the disk read could stay a block or two ahead of the sequential read through the buffer in RAM - most times 3-4 blocks worth of buffering was enough for each file being processed sequantially. ...
    (comp.lang.java.help)