Today on this ol' server

Monday May 22, 2006

JavaOne sessions

The best session I attended was Tim Bray's talk on Siggrid. Essentially, unless you're Google or Yahoo or Tim Bray, there are no service-oriented grids. He referenced Andy Bechtolsheim's HPC talk at Stanford. I was mucking last night trying to find a free player that reads wmv on a Mac. Stupid. Compatibility is not here, yet.

Went to the Java Technology in an Intelligent Swarm of Heterogeneous Lego Robots BOF. Although they weren't swarming as in boids, but a group of bots. They were pretty cool. Too bad the carpet piling sabotaged the front suspension. They'd react to movement and large stationary objects best. Skinny chair legs were hard for them, but to their credit I think they only hit the chair legs once.

The Sun SPOTs were in the Sun SPOT Sun Labs booth and a hands on lab. Mad props to Eric Arseneau, Christina Cifuentes and Derek White for pulling the lab together. They contacted me in late March saying it didn't run on Solaris. A few weeks later they had a Solaris driver. Even with a very rushed, last minute usb driver, we didn't have any machines crash. Also, Dan Price dropped by and showed me that on OpenSolaris x86/x64 you can enter the kernel debugger, but gnome won't release the screen, meaning that if you really know mdb, you can blindly interact with it. It was awesome to see a true expert fly through mdb. Technically inspiring, at a time when I was very burned out on logistics.

I stopped by the Building Highly Dynamic Battlefield Network Infrastructure for Boeing U.S. Army Future Combat Systems, using JXTA session. Think of a transmitter attached to each solider feeding information in real time back and forth, giving global information beyond a soldier's natural purview, sending and receiving information to each soldier. It reminded me of something like Massive (the AI program used in LOTR to program Orcs) overlaying a soldier but instead of Massive telling the AI unit where to go, it gives information to the soldier. All in realtime on JXTA, a network independent P2P protocol. The military always seems a big proponent of pushing technology. -sigh-

In working the hands on lab I met James Todd from the JXTA lab. JXTA is an awesome technology. Maybe it will fulfill a niche so we don't have to take our system administrators everywhere. Not that I don't mind traveling, it's just that you don't take your mechanic with you to Tahoe.

Lastly there was the DTrace tech session and hands on lab. I met Adam and Jarod a day or so before at the OpenSolaris booth. Awesome. Their talk was good fun. Maybe this is just me, but after fighting with the docs online, it's much easier to understand technology when true experts explain it. Either that or my day-to-day work is not giving me enough time to learn new technology. The DTrace hands on lab was about 90% full, which is great considering it was one of the last labs of show.

And of course the parties were plentiful. On Thurs I just needed some down time. After long weeks of build-up I just relaxed in our favorite little cafe in the city with some paper and pen.

be nice to yourself

In the downtime between a large show, I'd like everyone to take some time to be nice to yourself.[Read More]

Sunday May 21, 2006

JavaOne 06 Wrap-up

Too bad I didn't take pictures. Yesterday, late afternoon I said goodbye to Moscone and JavaOne 06. It's amazing how much work goes on behind the scenes to put on such a large venue. If only I had pictures of the break-down an set up. Amazing, the beeping, taping, and all the paper. Shouts out to everyone who worked the show. We did an amazing job.

I was the project lead system administrator for the Hands on Labs at JavaOne '06. For the tech specs, we had 100 single processor Ultra 20s with 2gb of ram, (10 machines had 4gb), running Solaris Nevada build 35, a multitude of Java versions, Netbeans versions, and the latest release of Netbeans 5.5 Enterprise Beta, Sun Grid engine, dual booted to SUSE 10 with Netbeans Mobility pac. On the morning of the 12th we physically brought up the 100 and deployed Suse, then Solaris to 90 brand new Ultra 20s. The other ten I had been using a mock environment about a month and a half. The server was a Sun Fire v1280 with 16gb of ram, and 8 cpus that was also used for last years show. This year we had a full gigE network, although it was not connected to the Internet. Sorry, to the lab users who couldn't reference the net for online documentation.

Zip files of the hands on labs can be found here: http://www.javapassion.com/handsonlabs/ That page will be updated with the latest zipfiles some time in the next two weeks (after I come back from vacation.)

Calendar

Feeds

Search

Links

Navigation

Referrers