On 8 Jul 2011, at 22:44, Sampsa Laine wrote:
Realistically what would it draw, compared to say an Alphaserver 4100 with two BA365s, fully populated?
I've got some space in a rack in my machine room, with air conditioning :)
I'm not sure about the AlphaServer, but A rx6000 with 2 Itanium 2 single core units draws 340-370 Watts and kicks out monster amounts of heat for a 2U.
If you want to effectively virtualise OpenVMS you'd be way better off running small x86_64 VMs with an Alpha or VAX emulator on a multi-core x86_64 server. Cost weighting of purchase vs. running costs makes way more sense.
--
Mark Benson
My Blog:
<http://markbenson.org/blog>
Follow me on Twitter:
http://twitter.com/mdbenson
"Never send a human to do a machine's job..."
On 8 Jul 2011, at 22:55, hvlems at zonnet.nl wrote:
My alphaserver 1200 with two 5/533 cpus, 4GB memory and 8 disks internally draws 500 Watts. That's 12 kWhr each day, around 2,5 per day.
A fully loaded 4100 has 4 cpus and draws as much as 650 W (swag).
My monthly power bill is 140 / month. No way I'll run systems 24/7 ever again.
An integrity is not an option I guess.
Hans
Verzonden vanaf mijn draadloze BlackBerry -toestel
Not for one user I suppose - but if someone wanted to run one, I'd rent a VM off it for a reasonable share of the operating costs..
Sampsa
My alphaserver 1200 with two 5/533 cpus, 4GB memory and 8 disks internally draws 500 Watts. That's 12 kWhr each day, around 2,5 per day.
A fully loaded 4100 has 4 cpus and draws as much as 650 W (swag).
My monthly power bill is 140 / month. No way I'll run systems 24/7 ever again.
An integrity is not an option I guess.
Hans
Verzonden vanaf mijn draadloze BlackBerry -toestel
-----Original Message-----
From: Sampsa Laine <sampsa at mac.com>
Sender: owner-hecnet at Update.UU.SE
Date: Fri, 08 Jul 2011 22:44:56
To: <hecnet at Update.UU.SE>
Reply-To: hecnet at Update.UU.SESubject: Re: [HECnet] House heater anyone?
Realistically what would it draw, compared to say an Alphaserver 4100 with two BA365s, fully populated?
I've got some space in a rack in my machine room, with air conditioning :)
Sampsa
On 8 Jul 2011, at 22:42, Mark Benson wrote:
On 8 Jul 2011, at 21:38, Sampsa Laine wrote:
On 8 Jul 2011, at 21:33, Joe Ferraro wrote:
The rx4640 should hold four dual-core I2s (1.6GHz / 24MB L3), and should have Madison / Tukwila availability to allow one to run openVMS under HP Integrity Virtual Machines.
That'd be kinda cool, if we stuffed it full of RAM we could be provisioning little virtual VMS 8.4 IA64 boxes to people on HECNET, no? Virtual DECNET hosting :)
Good luck raising the mortgage to pa y the electric bill for running hat monster 24/7. Most data centres would kill your wallet to host something like that because of the power and heat dissipation problems vs. the tweeny little dual-core 1U rack units most web hosting companies use. :)
It's a cool idea though all the same!
--
Mark Benson
My Blog:
<http://markbenson.org/blog>
Follow me on Twitter:
http://twitter.com/mdbenson
"Never send a human to do a machine's job..."
Realistically what would it draw, compared to say an Alphaserver 4100 with two BA365s, fully populated?
I've got some space in a rack in my machine room, with air conditioning :)
Sampsa
On 8 Jul 2011, at 22:42, Mark Benson wrote:
On 8 Jul 2011, at 21:38, Sampsa Laine wrote:
On 8 Jul 2011, at 21:33, Joe Ferraro wrote:
The rx4640 should hold four dual-core I2s (1.6GHz / 24MB L3), and should have Madison / Tukwila availability to allow one to run openVMS under HP Integrity Virtual Machines.
That'd be kinda cool, if we stuffed it full of RAM we could be provisioning little virtual VMS 8.4 IA64 boxes to people on HECNET, no? Virtual DECNET hosting :)
Good luck raising the mortgage to pa y the electric bill for running hat monster 24/7. Most data centres would kill your wallet to host something like that because of the power and heat dissipation problems vs. the tweeny little dual-core 1U rack units most web hosting companies use. :)
It's a cool idea though all the same!
--
Mark Benson
My Blog:
<http://markbenson.org/blog>
Follow me on Twitter:
http://twitter.com/mdbenson
"Never send a human to do a machine's job..."
On 8 Jul 2011, at 21:38, Sampsa Laine wrote:
On 8 Jul 2011, at 21:33, Joe Ferraro wrote:
The rx4640 should hold four dual-core I2s (1.6GHz / 24MB L3), and should have Madison / Tukwila availability to allow one to run openVMS under HP Integrity Virtual Machines.
That'd be kinda cool, if we stuffed it full of RAM we could be provisioning little virtual VMS 8.4 IA64 boxes to people on HECNET, no? Virtual DECNET hosting :)
Good luck raising the mortgage to pa y the electric bill for running hat monster 24/7. Most data centres would kill your wallet to host something like that because of the power and heat dissipation problems vs. the tweeny little dual-core 1U rack units most web hosting companies use. :)
It's a cool idea though all the same!
--
Mark Benson
My Blog:
<http://markbenson.org/blog>
Follow me on Twitter:
http://twitter.com/mdbenson
"Never send a human to do a machine's job..."
Somewhat related to the virtual VMS box provision, I've been playing with FreeAXP and have a Tru64 v5 box up now.
Let me know if anyone wants access..
Sampsa
On 8 Jul 2011, at 21:33, Joe Ferraro wrote:
The rx4640 should hold four dual-core I2s (1.6GHz / 24MB L3), and should have Madison / Tukwila availability to allow one to run openVMS under HP Integrity Virtual Machines.
That'd be kinda cool, if we stuffed it full of RAM we could be provisioning little virtual VMS 8.4 IA64 boxes to people on HECNET, no? Virtual DECNET hosting :)
Sampsa
The rx4640 is interesting as it makes use of so-called "cell"-architecture, unlike the single-
"cell" "entry-level" systems (like the rx1600/rx1620, zx6000/rx2600/rx2620 and so forth), thus
a bit more like the 'big iron' Integrity systems (like the rx7620, rx8620 and the Superdome).
I believe the rx4640 can take two, which can be fitted with four single-core processors each
or two "mx2" dual-processor modules (as seen in the first Integrity Superdome systems).
It certainly matters not; however, I believe myself to be correct in stating that cell architecture begins at the rx76 level and above.The non-cell rx6600, for example, is considered "entry level", is fashioned for the [mostly non-existent] HP/UX app-tier space due to the large number of onboard SAS drives (I have eight... I believe it will hold 16). Otherwise, a large number of onboard drives, as you've alluded, is pointless since HP/UX finds itself mostly within the database tier (rx7640, etc.. hold four onboard u320s). Boot from SAN with HP/UX is, again, rare (said "pointless"), since the complication that ensues has little (if any) return on investment excepting, perhaps, infrastructure based disaster recovery.
The rx4640 should hold four dual-core I2s (1.6GHz / 24MB L3), and should have Madison / Tukwila availability to allow one to run openVMS under HP Integrity Virtual Machines.
On 8 Jul 2011, at 15:29, Gregg Levine wrote:
On Fri, Jul 8, 2011 at 9:42 AM, Mark Wickens <mark at wickensonline.co.uk> wrote:
Just to remind you guys that the live feed is running from NASA for the last
space shuttle launch.
http://countdown.ksc.nasa.gov/shuttle/countdown/cdt/http://www.nasa.gov/multimedia/nasatv/index.html
Regards, Mark.
Hello!
Of course Mark. But the weather may cause it to be scrubbed, because
right now they are experiencing the worst weather in about seven
years. Which for Florida is a good thing. They need more rain then we
do.
You'll get a much better feed from www.spacevidcast.com in HD and you can chat to other space geeks while you are watching.
I'm an active spaceflight enthusiast, thanks to the power of Twitter. I went to Florida to see the launch of STS-134 (last flight of Endeavour) but missed it due to the scrub :( I know a ton of people on-site at KSC and around the area at the moment who are all waiting with baited breath.
Weather was quoted as 30% go but it's not looking too bad - might be a chance to launch today. If not it's improving over the weekend.
Oh and GO ATLANTIS!! :)
--
Mark Benson
My Blog:
<http://markbenson.org/blog>
Follow me on Twitter:
http://twitter.com/mdbenson
"Never send a human to do a machine's job..."
On 8-7-2011 12:10, Johnny Billquist wrote:
Maybe time to curb this discussion? It has nothing to do with HECnet, and it's becoming quite a few posts... Not everyone on this list even runs VMS, or cares about itaniums... :-)
Feel free to continue it in direct mails, or some other forum perhaps?
Okay, will do.
- MG
