Summer Institute 2004

Overview

From July 12-23, 2004, OSC Summer Institute participants experienced firsthand the dynamic fields of high performance computing (HPC) and networking. Working with kids their own age and with similar interests, they contemplated everything from computational chemistry to mechanical engineering, all while applying newly learned programming and visualization skills.

To make the experience well-rounded, OSC invited experts from the science, engineering, HPC and networking communities to speak with the students about relevant topics and careers in science and engineering.

SI is not just about spending time in the classroom. Students enjoyed a day of learning about team building while navigating obstacle and rope courses at the Adventure Education Center. Other fun field trips with a science spin included visits to OSU’s Nuclear Reactor Lab, Museum of Biological Diversity and The Center for Automotive Research. Students also got a taste of campus life by living and dining in the OSU dorms.

Participants

Boyan Alexandrov
Centennial High School

Chris Breneman
Cincinnati Country Day School

Chris Chang
Centerville High School

Chester Chen
Sycamore High School

Steven Dee
Garfield High School

Erich Kreutzer
Upper Arlington High School

Sean Lee
Thomas Worthington High School

Kevin Lin
Sycamore High School

Amy Liu
Hoover High School

Dylan Petonke
West Carrollton High School

Zach Sjostrom
Thomas Worthington High School

Richard Testani
Dublin Coffman High School

James Wang
Hoover High School

Norman Wang
William Mason High School

Kyle Wilson
Dublin Coffman High School

Projects

The students worked together in small teams on diverse and challenging projects. Teams were comprised of a project leader (staff member who conceived and designed the project), 3 or 4 students, a high school teacher and a student leader (in charge of dividing project tasks).

The projects offered change from year to year, This year's projects included: Parallel Processing, Electron Scattering and Fluid Dynamics.

The Parallel Processing project was designed mainly to teach the parallel processing language needed to split up a problem and assign different tasks to different processors. Their parallel code was run on supercomputers containing 128 and 256 separate processors. Two groups used parallel processing to make high-speed simulations in two different scientific areas. One group's projects studied the dynamics of stellar clusters. A cluster is a group of stars bound together by gravity. The movement of an individual star is determined by the gravitational force on it from all the other stars. Their code produced 3-D images of all the stars at times starting with the cluster formation. A processor was assigned to each of the 200 stars. Not only did the students use actual astronomical data from science journals to set up the cluster parameters but they made animations that looked amazingly close to the actual telescopic observations.
Click here for the animation.

The second parallel processing group studied the spread of an epidemic through a community. A different part of the community was assigned to each processor. The students performed a number of simulations each one on a different disease. They varied such disease parameters as contagious level, virulence level, transmission method, time to recover (if possible), and possible immunity. By changing these properties, the group simulated diseases that ran the gamut from the common cold to the Ebola virus. As with the star cluster project, actual data was used. This time obtained from the Center for Disease Control (CDC) in Atlanta.
Click here for the animation.

In the Electron Scattering projects, the groups considered a variation on Rutherford's theme, and developed a simulation of a beam of electrons bombarding a thin foil of some material. This basic process underlies the operation of a number of modern devices, including Scanning and Transmission Electron Microscopes (SEM and TEM). The basic idea behind these is to use the pattern of scattered electrons to reconstruct a picture of the object under study.

One team analyzed how the scattering would change if the energy of the incident radiation was changed. The animations produced showed that with higher energy, fewer particles "bounced back" and more when completely through the gold foil. The results were reversed for lower energy input beams. The track of each incident particle was colored coded to show how much energy that particle had at that moment in time.
Click here for the animation.

The other team kept the incident energy the same, but changed what type of material the particles. The simulated the electron scattering for low-density foils made of copper, for example, and high-density foils made of, say, lead. In the simulations it could be seen that the number of particles making it completely through the material was lower as the material density increased.
Click here for the animation.

The fifth group wrote code that simulated the difficult problem of Fluid Dynamics. When performing such simulations the characteristics of the fluid itself first have to be taken into account. Properties such as viscosity, compressibility, and density to name a few. Second, the interaction between the fluid and the material it is flowing through must be considered. A few of the questions to be answered included: Is the flow in a large or small conduit? Are there curves and bends in the pipe? What is the drag/friction between the fluid and the material's surface? This group produced a series of animations with different fluids and different pipe geometries. In several animations, clear constructive and destructive interference patterns were seen. All the data was calculated by solving the two/three dimensional partial differential Navier-Stokes equation.
Click here for the animation.

With the guidance of OSC faculty, students did their own work from code implementation to final results. The students took what they learned in the classroom and applied it to the coding and completion of their projects. An understanding of the project's science/engineering were prerequisites for facilitating code writing. The students made a video animation displaying their simulation data -- which was the ultimate goal of each project. Groups presented and explained their results to parents, OSC staff, and guests who attended the SI Closing Ceremonies.

Schedule

Click here to download a detailed schedule.