Monday January 21, 2008 I have had these pictures sitting on my laptop for about 1/2 a year. I pulled a couple out for a screen cast that I still haven't published because the audio blows. I have been doing a better job again on publishing pictures shortly after I take them. Network.com is the Sun Grid Compute Utility, I have a few pictures from the on site work.



Best Wow Blog:
Blog that is about CEC and leverages as many different technologies as possible to communicate its message. Chosen by a panel of judges, approved by the CEC Executive sponsor Dan Berg, will review blogs for:
- Wow factor: Does the blog incorporate technology in unique ways? Is it fun to read?
- Is this a blog that will last throughout the year?
Sweet, I have incorporated technology in an interesting and or unique way and I am fun to read.
Alec Muffett won best podcast for his “Terabyte iPod” presentation (which was quite interesting, you
should watch it) he is wondering if he can get a UK Wii.
Lou Springer won Best Red Shift Blog. Go Lou! I met Lou at CEC having previously conversed on
twitter and had dinner with him (and some other people see my dinner post listed below :) ) the last day of CEC.
I occasionally have a need to find queues in Sun Grid Engine that are in one of the possibly problematic states which have an occupied slot. It is just infrequent enough that I don't remember exactly how I did it the last time.
qstat -f | awk '$6~/[cdsuE]/ && $3!~/^[0]/'
queuename qtype used/tot. load_avg arch states
zone.q@r130c24z0.network.com BIP 1/1 -NA- sol-amd64 adu
zone.q@r130c24z1.network.com BIP 1/1 -NA- sol-amd64 adu
An alternate is "qstat -f | awk '$6~/[cdsuE]/ && $3~/^[1-9]/'" which also avoids printing the header line. In the example above 'state' in $6 matches 's' and 'used' does not begin with '0'.
The possibly more elegant 'qstat -f -qs cdsuE' still requires a second comparison in awk of '$0!~/--/' to filter out the queue separator lines. (qstat -f -qs acduE | awk '$0!~/--/ && $3!~/^[0]/')
Finally because I can never remember what exactly all the queue states are and the qstat man page doesn't have the nice table:
aoACD – Number of queue instances that are in at least one of the following states:
a – Load threshold alarm
o – Orphaned
A – Suspend threshold alarm
C – Suspended by calendar
D – Disabled by calendar
cdsuE – Number of queue instances that are in at least one of the following states:
c – Configuration ambiguous
d – Disabled
s – Suspended
u – Unknown
E – Error
Job State/Status:
d(eletion), E(rror), h(old), r(unning), R(estarted), s(uspended), S(uspended), t(ransfering), T(hreshold) or w(aiting).
References: SGE (N1GE 6.0) -- Monitoring and Controlling Queues
Edit: Added Job Status, literally couldn't find that in any of the online docs (notwithstanding ~40% through the qstat(1) man page, targeted google searches do a poor job finding the link)
I just sent Paul Monday over at the Blackbox Blog a link to the full gallery of pictures that I posted from our visit to the Blackbox while it was in DC. For your viewing pleasure I will post a few of them here.
As we pulled into the Wardman
Park Marriot I was suprised to
see a line of taxi's coming down the street. I had attended LISA
here a few months ago and the taxi line was never that long. I knew
that the Blackbox was popular but that was a bit more than I was
expecting.
When we rounded the corner it was a cool thing to see the
Blackbox Tour Mobile...taking up the middle of the normal taxi lane.
One of the taxi drivers came over right before we left and
asked C- what was going on. She informed him that this was Sun's
Project Blackbox... A whole bunch of computers on the back of a truck.
I don't know if he went on the tour.
Paul
just posted an interview
with Joe Carvalho, here is a
picture of Joe (or his hands) modeling a SunSPOT accelerometer.

We also got some swag from the event. As if the Blackbox itself wasn't cool enough we are the only people on our block who have the incredibly wrinkled Blackbox T-shirts and Trucks!
The T comes shrink wrapped and I guess crammed or vacuformed
into a truck shaped brick. The truck comes in a relatively
uninteresting box, of so little interest I didn't even post a picture
of box with the truck in it.
The Truck/Shirt still in Truck form.
The truck, I still think it needs to have racks in the back be a transformer...or both!

To end for now I leave you with a picture of the Emerency Power Off button. Which I did NOT press. I did ask if anyone had yet pressed it while on tour and was informed that the answer was no. Given that the demo is not a production environment I was me sorely tempted to press the jolly candy-like button.

I have been listening to/watching the Sun internal Blade Summit. (Really a couple of days ago, but I never got around to posting this)
Interesting notes on hardware in general. The things that struck me the most about the general hardware discussion was the power cost of FB-DIMMS. The expectation is that with the high power consumption of FB-DIMMS that the memory will use more, even significantly more power than the CPUs. We have more IO and higher CPU density. That even with the memory power consumption and our higher slot density we still come in at lower power consumption rates. Everything is fully hot swap and hot plug.
Of course there are an incredible number of other cool things that are available now and that will be available soon. Unfortunately as I am attending this because I am interested and I have customers that may be deploying full chassis in relatively large numbers. I can't quite justify not doing everything else I needed to do today.
Had I been paying more attention to the whole thing I would have a better feel about which of the really cool things I can freely talk about.
Mean Time Service Interruption (MTSI) is reduced, fewer components == fewer failures. Fewer higher reliability fans and power supplies etc.
The concerns aren't really any different. Cooling capacity, Floor Space, Power. High density in blades helps address floor space. Cooling and Power are the new black.
Peter Snelling, just gave a very interesting and compelling talk about the limitations caused by power draw and as a result cooling capacity.
I had to drop off again, but not before one of the parts I had been waiting for. The management aspects and how the Chassis and Blade management would interact. Not much of a surprise, now I just want to get my hands on a few and play.

Technorati Tags: secondlife, SUN, web2.0