The offcial video for the 2009 Beowulf Bash in Portland has hit the Internet. Check it out. Maybe you will see yourself or someone you know or someone you don't know and are glad you don't.

Of course I would be remiss if I did not mention the sponsors on last time; AMD, Econnectix, InsideHPC, Penguin, SICORP, Terascala, Xand Marketing, and of course ClusterMonkey.

When I travel to the west coast (I live on the east coast) it takes my body 3-4 days to adjust, which is always when it is time to head home. I was out last night for a late dinner, got back to the hotel about midnight and woke up wide awake at 5:30am. By this afternoon I will be dragging my feet. It may be time for my first Red Bull (maybe that should be plural).

I'm still shifting through the news and I'll have more coverage throught week, but the big "tend" seems to be GP-GPU computing. Everyone is talking about it and there are plenty of products on the show floor. Of course it could be that the whole concept of racking, stacking, and managing clusters has been discussed to the Nth degree and GP-GPUs are fresh idea worth considering.

The Beobash was a great success. Here are some pictures and comments from John Leidel over at InsideHPC. Thanks to all the sponsors and the Linux Chick for all their support. I'll be posting video next week. Speaking of video, I have been spending a bunch of time doing video un-professional interviews for Linux Magazine. They should be hitting the web next week as well.

Someone also said they saw a hard copy of "HPC Dummies" at the show. Boy it would be nice to get copy of the book I wrote (some may call it a pamphlet). I'll down a few Red Bull and go on a quest today to see if I can snag a copy. There is nothing like a caffeine crazed overly tired writer pestering you at a trade show.

One final bit of news. I have a prototype of my Limulus personal cluster workstation up and running. If you are SC09, it is in the SICORP booth (1209). You can find more information (pictures and slide deck) on the project site. You want one of these.

Ahhh, the pre-SC (supercomputing show) pull your hair out frenzy. I'll have lots more to announce about the show in the next few days. I wanted announce this years Beowulf Bash first because it is one of the best events of the show. Click the got beer button for the witty and clever invitation (cough, cough).

This year a local HPC company who wishes to remain anonymous has donated five kegs of custom brewed beer for the event. The party begins at 9PM on Monday November 16th, 2009 (after the opening gala).

We are hodling it at the The Game, at the Rose Quarter, One Center Court (one really long block + two short blocks from the convention center, at the Rose Quarter Max Stop). Click Here to see the map. For public transportation information, visit Tri-Met's website.

This years sponsors are AMD, Econnectix, InsideHPC, Penguin, SICORP, Terascala, Xand Marketing, and of course ClusterMonkey. See you there.

Ethernet has been a key component in HPC clustering since the beginning. Over the years, interconnects like Myrinet and InfiniBand (and some others) have replaced Ethernet as the main compute interconnect largely due to better performance. High performance interconnects like InfiniBand are now the interconnect of choice for those that require performance and scalability.

With the availability of such high performance interconnects, one has to ask why do people still use Ethernet? The answer is three fold. First, Gigabit Ethernet (1000 Megabits/second) is "everywhere." Multiple Gigabit Ethernet (GigE) ports can be found on almost every server motherboard and users are comfortable with Ethernet technology. Second, it is inexpensive. The commodity market has pushed prices to the point where low node count clusters can expect Cat 5e cabling and switching costs to be between $10 and $20 per port. And finally, Ethernet is virtually plug-and-play. In other words, it just works.