Cool, I remembered my advogato password ;-)

I am getting married in less than two weeks! My wife-to-be has kicked my butt to make progress on my Ph.D... Yes, I am STILL working on my Ph.D. I passed an important milestone a month ago, and I now see light at the end of the tunnel.

A quick open source software update... Last year I became a co-developer on the Warewulf cluster project. If you do or are thinking about doing beowulf cluster work, check it out.

Back to the grindstone... I'll post again in a year or so I guess ;-)

Well, it's been awhile since I've written anything here... over 3 years! And, yeah, I'm still working on my Ph.D. at UK (U. of KY). I'm at my sister's in Atlanta for the holidays.

The FNN stuff we (myself and my advisor, Hank) started developing in 2000 has lead to all sorts of stuff. Too much to go into right now. As a side note, that little $10K cluster from Oct 2000 was called the KRAA Z-MP and actually turned out to be a real pain... Someday I might write up an article about what not to do when building a PC cluster... It's been decomisioned and several of the nodes are now workstations in one of our labs.

The cluster worth checking out is our new 128+4 node creation KASY0 which we built this summer.

Hmmm, too much going on to post it all...

I'm presenting a paper at ALS2000 on Thursday morning. I still have some tweaking to do on my slides, but I've already got a full set done... just too many of 'em. My flight to Atlanta is in 22 hours.

SC2000 is approaching rapidly, and there is a LOT left to do before our research group will be ready to go. We are building a new cluster (yes, another one) for under $10K for the HPC.Games contest. It is going to be sweet when it's finished. However... Some PC-Parts vendors really annoy me... out of five 256MB PC133 DIMMs delivered from someplace that will remain anonymous, only one survived more than 30 minutes with MemTest86 without errors. Don't companies do burnin testing of anything anymore?

Well, that's all for now.

Hmmm, I seem to have been demoted from Apprentice to Observer. Not that I'd used my former status as Apprentice to make any article postings... I still find it odd, since I have several certifications as apprentice from others who are certified as apprentice or higher. Oh well for now.

On another note, I just spent about 2 or 3 hours reading some of the recent articles and discussions here on Advogato. I can remember when Slashdot was only half as good, and it's only gone downhill from there. Advogato sure has some very intelligent discussions relating to software development. I've picked up a few really good references to some tools and techniques that I hope to use in AFAPI related things.

Our SC2000 paper was revised, and hopefully will now be readable by most users of the proceedings later this year. We had a tough time getting a fully working Postscript document to covert to PDF without introducing errors. Computers are so advanced, yet it is still a major problem to distribute documents electronically in a way that is truely portable across platforms/OS. My advisor's solution of using troff works for him, but our collaborators on this project use LaTeX, and well, the two don't mix well. I guess it's a universal problem: the perfect "versioning control system", "project configure and make facility", "programming language", or "portable document format" just don't exist yet, and probably can't due to some yet undiscovered law of nature :-).

The UKLUG group has gotten a small subset of their machines set up and configured as a video wall with AFAPI and VWLib. For now (and probably for quite some time), their stuff is housed in the KAOS Lab. This has prompted me to look into some unfinished business with the merge of VWLib into the AFAPI distribution. But classes start this Wednesday, so who knows when I will get a chance to resolve the issues with the merger. I've got other higher priority things to work on.

Speaking of higher priority things, Hank & I are working this week on our final version of a paper to be presented at the Annual Linux Showcase in Atlanta on October 12. It's "yet another paper" about KLAT2 and it's FNN.

Hmmm, what was I saying about having time to try doing an AltiVec version of our LINPACK optimizations??? Oh well. Anyone reading this know how to make human clones? I need two or three more of me :-)

Oh yeah, that takes funding... maybe next month :-)

Ack, what happened to July?

• Hank and I hacked with Thomas on his CFD code on KLAT2 to get final performance numbers for the Gordon Bell prize entry.
• Wasted a week with a misunderstanding if the -ARP flag for /sbin/ifconfig. Long story short: -ARP turnes off everything related to ARP for that NIC, including use of any preloaded/constant entries in the ARP cache! What a useless option.
• Hacked on the GA for finding/optimizing FNNs. We now have a better framework for specifying specific communication patterns for the GA to optimize for. Still not much closer to releasing the source :-(
• I took a week vacation to attended MacWorld Expo 2000 with my Dad. I promptly drooled on a G4 Cube. In a way, I wish I didn't already have a G4 Tower!
• Upgraded Odie to gigahertz Athlons and ABIT KA7 motherboards that AMD donated. They are fast, but my 450 MHz G4 still cranks out a few more RC5 keys/sec.
• Upgraded Opus to 500 MHz K6-2 CPUs that AMD donated. This made a very perceptable difference in the speed of the video wall, so I guess VWLib isn't memory bandwidth limited...
• Built and tested one of the AFN000601 boards. Will need to recruit some students to help assemble 20 more to use on KLAT2. Still waiting on one critical part (74LS13 chips) to be delivered.
• The University of Louisville's S+LUG visted the KAOS Lab to get a tour and some assistance with their cluster project.
• The University of Kentucky's UKLUG has begun work on converting a pile of 486's in the KAOS Lab into a usable cluster (Galugtica) for learning/hacking purposes.

So, it was a busy month of July for me. The fall semester is rapidly approaching. But there are still a few weeks left of "no classes". MacWorld Expo 2000 was a fun vacation to take with my Dad (and Mom). While at MacWorld, I met and chatted extensively with Troy Benjegerdes about AltiVec, BlackLab/ YellowDog Linux and parallel computing. Now that the Gordon Bell paper is done, I might have time to try our LINPACK optimizations out on my G4 with AltiVec.