Category Archives: Announcements

Dr. H.J. Siegel, an old colleague of my advisor and a leading figure in multiprocessing, will be at UK to give a talk next Monday. Abstract:

Title: An Introduction to Research Issues in Heterogeneous Parallel and Distributed Computing

Time: 10AM, Monday, March 1, 2010

Room: 112 RMB

Abstract:

In heterogeneous parallel and distributed computing environments, a suite of different machines is interconnected to provide a variety of computational capabilities. These capabilities are used to execute a collection of tasks with diverse computational requirements. The execution times of a task may vary from one machine to the next, and tasks must share the computing and communication resources of the system. An important research problem for heterogeneous computing is how assign tasks to machines and schedule the order of their execution to maximize some given performance criterion. An overview of a conceptual model of what this involves will be given. An example of resource allocation research will be presented. The example involves an ad hoc grid environment, with energy constrained mobile computing devices that could be used in a disaster management scenario. Open problems in the field of heterogeneous parallel and distributed computing will be discussed.
“Alligators” that make heterogeneous computing challenging will be shown.

To give a little background on Dr. Siegel, tweaked from a discussion with a friend and former group-mate who went to do his PhD under Dr. Siegel:
“There is a threshold, somewhere around tenure in academia, at which you are no longer effectively discouraged from cultivating eccentricities.” This is true for most tenured faculty, but the cowboy + leading figure in multiprocessing combination is a winner among winners.

My classes have met for the first time this semester, so it is time for my customary class impressions post. Older similar posts are archived, the link chain starts here. I am only taking 6 hours this semester, originally because my preferred last core class was not offered. I was thinking this would allow me to get ahead on research this semester, but is now also fortunate because of the TA position. The lab sections don’t start meeting until next Tuesday, so I can’t start talking shit (up to FERPA-approved limits) about that yet.

CS585:Linux Internals/Finkel
There are a lot of familiar people in this class, instructor included, and I’m pretty sure it is going to be awesome. It seems like the class is going to be exactly what I hoped; we’re going to dive into the kernel sources and hack around, guided by the books and lectures. We’re using two books, Linux Kernel Development, which is written in the fabulous tongue-in-cheek manner that seems to be endemic to good computer scientists, and Understanding the Linux Kernel, which is an O’Reilly book in the standard tradition.
This should more than make up for the extremely lackluster undergraduate OS class (CS470) I had from UK, seeing as we basically covered CS470, less some tedious detail on implementing resource locks and using shared memory, in the first lecture. Very excited, and expecting very hard projects.

PSY562:Human Technology Interaction/ Carswell

We did the around-the-room introductions thing, and the composition of the class should make things really interesting: 11 Psychology Seniors, a Computer Science Senior, a Computer Science Masters Student, an Education Graduate(didn’t catch which) Student, an Information Sciences Masters Student, a Computer Security person, and myself. In addition to the professional variety, we have hobbies like “Professional Juggler,” “Snake Breeder,” “Dog Trainer,” and “Certified Skydiver,” so there should be no shortage of interesting people. I suspect groups will always be set up with the topics people evenly distributed among the psych kids for the mutual exposure. Apparently the class is going to be fairly guided, and run around a selected central project (”Something significant to the Lexington community”, but we don’t know what yet), which means there won’t be quite as much chance for implementation as I would have liked, but it should be great fun anyway.

There is one other person in both these classes, which proves(or at least allows me to pretend) they aren’t a totally irrational pairing of interests. Looking forward to the semester. Good skills to be had, and it looks as though the classes themselves will be fun.

My big new behavior for the coming semester is that I’ll be working as a teaching assistant (for those unfamiliar with the system: at the university level TAs are graduate students who handle some portion of the instruction for classes). It had been mentioned to me several months ago that I might be asked to TA EE281 (the basic digital logic lab at UK) this semester, but because no one ever got back to me, I thought they had found a PhD student to do it. Last Tuesday I got a note that in fact it hadn’t been resolved, and, not knowing it had been previously mentioned to me, asking I would be willing to TA. It isn’t ideal; It ruins my perfect “Class only on T/R afternoon” schedule in a big way, one of the lab sections has an awkward little overlap with one of my classes (which I’m told will be worked around), and handling 3 lab sections will be a fair amount of work but I agreed anyway. Its good for the CV, its good departmental sucking up, it pays significantly more than my living expenses, and most importantly, its an experience I would like to have early on since I’m seriously looking at a career in academia. Monday I find out exactly what I’ll be doing from the faculty member running the course; I suspect I will be handling the lab sections and most of the grading, but not much of the lectures/lab design, which should be good for easing in. I might see about doing a little bit of the other parts as well just for the experience.
Before you are allowed to TA, the university requires an orientation. Said orientation took up Thursday and Friday, 8:30A-4:00p, far longer than was actually needed for the amount of content, despite being shorter than the summer version. I’m glad for the orientation, and learned a lot of valuable information, but some of the material the first day was pretty bad; things with as much BS in them as the ed-psych people usually only say “moo.” The good stuff from the first day included the obmbud information session (ass-covering rules, syllabus constraints, etc.), and we had a fairly solid setup for the microteaching exercise. The second day started two hours late due to weather, and was improved for it. The morning was a quick procession of presentations on rules, regulations, recommendations, and resources both for TAs to use ourselves and to refer students to, which did include the joke worthy “Rules about sleeping with your students.” The latter half of the second day was spent running and group critiquing our microteach lessons, and everyone in my room did quite well, definitely better than some of the instructors I’ve had. For my session I ran a 7-minute math-free version of the “Know your parts” style lesson on Light Emitting Diodes, at the hobbyist/beginning EE student level. Even the people who didn’t have the background to completely follow seemed to think my technique was good, and I’m not too embarrassed watching the video now (Things I see now that no one commented on: I made a few dumb omissions to keep time, and looked at my note page too often), so I probably did a reasonable job. I’ll grant that the microteaching system is a good way of vetting and improving teaching ability, and I tend to be quite skeptical of meta-education.
Some of the anecdotal content was pretty useful as well. There were TWO versions of lessons on not being bullied by football players and other large entitled people, because it is apparently that much of an issue. It was also interesting to hear from people already TAing; one of the existing TAs who has been teaching Chemistry 105, which is gigantic and incredibly hostile (its the scraper course kids who mistakenly think they are pre-med), had a lot of good questions on classroom management, particularly with regard to cheating and hissy fits by students that brought up lots of useful information.
As an additional source of irrational excitement, I get one of those nifty blue-flap UK Graduate School messenger bags, which I’ve always really liked for some reason.

I just submitted my last project/paper/assignment for the semester, and it feels great. In talking to instructors and fellow students, it sounds like the principle of least-fuckup has asserted itself again, so my grades should be fine even though several of the things turned in in the last few days aren’t as good as I would have liked. Now I just have to attend class tomorrow, and remember to go to my one final next Friday, and all will be well.