Visit to Army High-Performance Computing Research Center and the Twin Cities Campus

Figure 1: Bess Vljaisavljevitch, Chris Orth, Tyler Hutchison, Daniel Selifonov, Scott Lembcke,

Andy Korth, Arthur Aaberg, Anna Schliep, Sam Johnston,

Mike Anderson,  Matt Gravelle, and Elena Machkasova.

 

On March 23, 2006, a group of 11 students and 2 faculty took a trip to the Twin Cities in order to visit several landmarks in computer modeling. A variety of majors were represented among the students, including physics, computer science, mathematics, chemistry and biology, the faculty were Elena Machkasova from Computer Science, and Sylke Boyd from the physics discipline. Originally planned as a class field trip for the course on Computer Modeling of Materials, the proposed trip quickly gathered interest among science students with an interest in computational modeling and high-performance computing.

 

Figure 2: Dr. Frances Hill explains the inner construction of an old Cray computer.

 

Dr Frances Hill, a staff scientist at the Army High-Performance Computing Center in Minneapolis, organized an interesting three hours at the institution. The visualization lab impressed us with a large screen, used in conjunction with motion recognition technology. Students could delve into a virtual 3d world which would adjust perspective according to their positions. Scientists from the center gave presentations on interesting modeling projects. These included the various applications of fluid dynamics using an adaptive-mesh finite-element approach. There were applications ranging from the animated airflow around various aircraft, including a large cargo airplane and helicopter, to the fluid flow through an artificial heart. Another direction of modeling took us into the world of projectile – solid interactions. These models involve materials under high strain to the point of fragmentation. Also, we learned about projects in computational chemistry.

Figure 3: In the machine room at the AHPCRC

Figure 4: The computing center from the outside. The big fans on the right part of the building provide cooling for the machine room.

 

 

The AHPCH features several top-notch Cray parallel computers in addition to an array of Beowulf clusters. On a tour of the machine room, we learned about the elaborate cooling mechanisms for these computers, as well as various facts such as the energy consumption of the center being equivalent to that of a small town. The center offers summer research programs for undergraduate students – these should pose an excellent opportunity for computer-interested students to learn about FORTRAN programming, programming of parallel computers and physical modeling. We appreciate the time that was taken to show us around and demonstrate the remarkably complex computer models to us. During lunch, there was opportunity for some chatting with the staff scientists at the center.

 

Figure 5: After lunch (sponsored by the AHPCRC) we crossed Washington Bridge to the University of Minnesota Twin Cities campus.

 

Figure 6: The group that visited the Department of Chemistry, UTC.

Bess Vljaisavljevitch, Anna Schliep, Arthur Aaberg, Chris Orth, Matthew Gravelle, and Mike Anderson.

 

One group visited the computer science department and learned about options for graduate school in computer science. This group listened to several presentations from graduate students, and visited the graphics lab, the robotics lab and a lab for research on artificial intelligence.

 

Another group visited with research groups in computational chemistry. Dr. Jiali Gao was so kind to welcome us for a tour of four research groups in the Minnesota Computational Chemistry (MC2) group. We met with professors, postdocs, graduate students and undergraduate students who are involved in a wide variety of computational research projects. Special thanks go to Prof. Jiali Gao, Dr. Ben Gherman (group of Prof. Chris Cramer), Prof. Darren York, and Prof. Ilja Siepmann, who took the time and effort to talk to us about research projects in their groups. The computational models deal with systems such as enzyme reactions, viral RNA reactions within the cytoplasm, neuro-transmitter oxidation, phase separation in gas chromatography, defect migration between planetary mantle and core, nucleation of water droplets in the high atmosphere, and others. We learned about the methods used in the physical models, such as the force fields, quantum-mechanical approximations, molecular-dynamics and Monte-Carlo techniques. Computer models can try to assess situations which are experimentally not feasible, such as high-pressure high-temperature systems, or very instable or inaccessible systems. We spent an intense three-and –half hours in the Department of Chemistry, filled with information, personal contacts and –last not least – an impression on what graduate school in computational modeling might entail. Two UMM graduates had gone through Prof. Ilja Siepmann’s group before: Collin Wick, who is currently a postdoctoral fellow at Pacific Northwest National Laboratory, and John Stubbs, who is presently a visiting assistant Professor at Grinnell College.

 

This visit was worthwhile to the students for many reasons: It was an opportunity to see how far the methods learned in various courses can be taken, to the most complex systems. It posed an opportunity to learn about real applications in research of high-performance computing and to learn about the implications (such as the prevalence of Fortran programming in scientific computing) relevant for computer science. For any student considering graduate school, it gave an opportunity to find out on how life at that stage might look like. For everybody, the trip provided inspiration and new contacts. Personally, I am very glad to have met faculty at the Twin Cities campus, and hope to foster the connections.

 

After a nice walk (that would have taken us through all of Morris and then some), we sat at a well-deserved dinner in the Old Spaghetti Factory.

Figure 7

Morris, March 25, 2006

Sylke Boyd

The views and opinions expressed in this page are strictly those of the page author. The contents of this page have not been reviewed or approved by the University of Minnesota.