Lecture Evaluation of the GPU Programming for Molecular Modeling Workshop held at the University of Illinois, Urbana

August 2-4, 2013

Questionnaire: Gila Budescu, TCB group, UIUC, and modified by David Brandon, TCB Group, UIUC
Analysis and report: David Brandon, TCB group, UIUC

The Theoretical and Computational Biophysics Group (TCBG), an NIH Center for Macromolecular Modeling and Bioinformatics is headed by Klaus Schulten, and CO-PIs Z. Luthey-Schulten,  L. Kale, A. Aksimentiev, and E. Tajkhorshid.  As part of its outreach, the Center offers workshops to introduce and transfer its programs and technological solutions to the biomedical community.  The Center organized a 2.25 day (August 2-4, 2013) workshop held at the Beckman Institute for Advanced Science and Technology on the campus of the University of Illinois at Urbana-Champaign.

TCBG development staff members John Stone, James Phillips, and David Hardy provided lectures, along with Fernanda Foertter, from the Oak Ridge Leadership Computing Facility, and Ivan Ufimtsev, from Stanford University. The program of the workshop consisted of lectures, participant presentations, and a programming project laboratory. Participants worked on their own laptops during the workshop, but also accessed supercomputing resources. The lecture evaluation form can be found here.

Lecture evaluation are comprised of two elements, 1) the proportion rating the relevance of the lecture or tutorial as highly relevant (i.e. 'very good' + 'excellent' ratings; see Table 1: Summary of Relevance Statistics below), and 2) select comments considered illustrative of respondent opinion.  As is frequently the case with surveys, not all respondents answered all questions; the number of responses for the relevance ratings (r=) and comments (c=) are listed next to the name of each lecture and tutorial summary, e.g. (N: r=12, c=12).

Some issues to consider when reading the comments:

  • Written comments, particularly when comments are extreme in one direction or the other, tend to stick in one's head more so than statistics that may present a more accurate summary of opinion.
  • Attendees appear to have been somewhat heterogeneous in computational background, training, interests, and experience; so, for any lecture or tutorial there was likely always someone new to the topic who needed more time, help and explanation, and at the same time someone very experienced who wanted more breadth and/or depth on the topic.

Day 1 Lecture: Heterogenous Computing on Titan (Fernanda Foertter, Oak Ridge Leadership Computing Facility) (N: r=11, c=11)

All participants (100%) rated the relevance of the lecture as very good or excellent.

  • Excellent talk. Good info on Titan.
  • Lecture is clear and easy to understand. Presentation provided useful information about applications on Titan.

Day 1 Lecture: GPU Particle-Particle Algorithms: Non-bonded Force Calculation (David Hardy, TCBG) (N: r=11, c=9)

All participants (100%), found the lecture content relevant.

  • Excellent talk, I thoroughly enjoyed the heavily technical aspects of this talk and the extensive discussion of GPU implementation techniques.
  • Most related talk to our project. Learned how David went over the problems he faced and how he implemented solutions. Gave me some ideas I may implement my code improvements.

Day 1 Lecture: GPU-Accelerated Quantum Chemistry (Ivan Ufimtsev, Stanford University) (N: r=11, c=8)

A majority of participants (73%), found the lecture content relevant.

  • Excellent, Ivan did an excellent job laying out the problem and Epiphany which was a very intriguing to see a new piece of hardware.
  • Cool hardware and good talk that showed us some new hardware.

Day 2 Lecture: Lattice Microbes: CUDA Algorithms for Stochastic Simulation of Biochemical Reactions (Mike Hallock / Joseph Peterson, U. Illinois) (N: r=11, c=7)

Nearly all participants (91%) perceived the relevance of the lecture content as very good or excellent.

  • I really enjoyed it. It was not only informative but also fun/funny. There were a good team.
  • One of my favorite talks in that I was able to most closely identify with it. Thoroughly enjoyed algorithmic design discussion.

Day 2 Lecture: NAMD: GPU-Accelerated Petascale Molecular Dynamics Simulations on Titan and Blue Waters (Jim Phillips, TCBG) (N: r=11, c=10)

All participants (100%) rated the lecture topics as very good or excellent.

  • Great knowledge…good communicator.
  • Very detailed lecture about NAMD. The way to overlap CPU with GPU is interesting.

Day 2 Lecture: GPU Particle-Grid Algorithms: Molecular Surfaces and Synthetic Density Maps (John Stone, TCBG) (N: r=11, c=10)

All participants (100%) rated the lecture topics as very good or excellent.

  • Great job! Answered a lot of outstanding questions I had re: vmd. Presentation was very straight forward and well delivered. Really was intriguing from an optimization perspective for molecule codes.
  • Very good, covered all the detailed skills and tricks how to utilize GPU. Excellent very knowledgeable get answers to questions even not directly related to the topic of the talk.

The complete set of comments is available by e-mailing workshop+gpuaug2013@ks.uiuc.edu.


Table 1: Summary of Relevance Statistics

                                                                                                
  N Poor Fair Good Very Good Excellent
Day 1 Lecture: Heterogenous Computing on Titan 11       55% 45%
Day 1 Lecture: GPU Particle-Particle Algorithms: Non-bonded Force Calculation 11       55% 45%
Day 1 Lecture: GPU-Accelerated Quantum Chemistry 11   9% 18% 36% 36%
Day 2 Lecture: Lattice Microbes: CUDA Algorithms for Stochastic Simulation of Biochemical Reactions 11     9% 36% 55%
Day 2 Lecture: NAMD: GPU-Accelerated Petascale Molecular Dynamics Simulations on Titan and Blue Waters 11       18% 82%
Day 2 Lecture: GPU Particle-Grid Algorithms: Molecular Surfaces and Synthetic Density Maps 11         100%