The Project

The goal of this team's project was to create a physically realistic Direct Simulation Monte Carlo (DSMC) model and optimize its performance on multi-core processors, making intricate simulation available on desktop computers versus a “supercomputer” that may cost tens of millions of dollars. The research could open doors to simpler methods of simulation of physical systems; such systems include the weather, re-entry of space vehicles (which this project studied), auto collisions and even modeling of complex biological processes such as molecular docking.

About Erika

Erika DeBenedictis, a junior at Albuquerque Academy, enjoys Physics the most because it allows her to not only understand what something is, but also how it works. She would like to pursue an undergraduate degree in Aerospace or Physics to satisfy her interests in the field. Ms. DeBenedictis thrives on independent research projects and focuses her work on a school science fair as an individual competitor and on the New Mexico Supercomputing Challenge as part of a team each year. Her hobbies center on music as she sings with the Girls Ensemble and her school's audition choir. She also enjoys piano. Her dream job would be to work on Aerospace research for Mars missions and beyond.

About Tony

Tony Huang, a senior at La Cueva High School, would like to continue on to major in Electrical Engineering for his undergraduate degree. He has participated in many science competitions through the years, and among his credits are honors at the Science Olympiads, Science Bowl/Quiz Bowls and MATHCOUNTS, where he also acts as an Assistant Coach. Mr. Huang is a member of the National Honor Society, and is actively involved in a mentorship at the Center for High Tech Materials. He would like to become a professor one day, and is interested in Aerospace and Engineering because of the admiration for his dad, an Air Force Research Lab Senior Physicist. He was born in Shanghai, China where he lived until he was three.

Text courtesy of the Siemens Foundation

More About All the Siemens Competition Finalists

The Siemens website has webcasts of all of the finalists' presentations and of the awards ceremony.

.::: Q & A with Erika Debenedictis and Tony Huang :::.

What was your project about and how much actual work did you do?

Erika: Our project is looking at how to accurately and efficiently simulate movement of objects in the upper atmosphere. To do this we implemented a mathematical model called Direct Simulation Monte Carlo (DSMC), which accounts for the unusual behavior of low-density air.

While DSMC is a way of modeling air flow that has already been established, Tony and I did some really innovative work while trying to implement it. First, we had to convince ourselves that the model was working, so we researched and developed various ways of testing the program, including checking if a box moving at different supersonic speeds forms the proper shockwave shape.

The most interesting part of our project came from trying to get the simulation to run efficiently on multi-core processors. Rather to our surprise, we found that even when we were using every bit of processing time the program didn't speed up as much as it should have. A few more tests showed this was caused by a bottleneck in how fast all the processors could access their memory. Since multi-core processors are so new, this had never been observed with DSMC before, and was a striking example of the limitations of the multi-core architecture.

Could you explain the project in layperson's terms and how it could help things in the real world? The description listed is kind of abstract.

Tony: Our project looks at a computer algorithm that simulates the upper atmosphere, and in particular, objects such as space shuttles moving through the upper atmosphere. The project has many potential benefits because we can use the model to analyze things like drag and temperature of the spacecraft. In our model, we can also view a density map of the air surrounding the space shuttle. This information could possibly used to analyze the progression of a space shuttle as it re-enters the atmosphere. Any piece of reusable space equipment will eventually need to re-enter the atmosphere. Therefore, it is important for scientists to simulate the re-entry extremely accurately. A slight error could mean the difference between the space shuttle landing safely or a thousand miles off target.

How did you choose your project?

Erika: Tony and I spent two or three weeks brainstorming ideas and researching them before we decided for sure on a project. We started out with about ten or so pretty vague ideas and narrowed it down to just a couple more complete project plans. We actually convinced ourselves we'd do a project on de-blurring images, and then we researched all the top project ideas more and realized DSMC would be much more fun.

What kind of work did you do as scientists, besides asking questions, doing experiments, etc.?

Tony: As scientists, we spend a lot of time just looking for things that arouse our interest. Whether it is simply surfing news sites on the Internet or actually reading scientific articles, it is important to keep one's mind open for areas of possible new research. When it is actually time to do research and put together a project, we spend a lot of time not only looking for information, but also making connections.

What did you and didn't you include in your research report?

Erika: When we decided to enter Siemens Competition we had a 34 page 1.5 spaced report, and needed to cut it down to 20 pages 2.0 spaced. Painful as it was, the report came out SO much better because we were space limited. We had an overview of what DSMC is and how it works, a description of the simulations, and a section on the multicore memory bottleneck.

So what got cut out? Mostly descriptions of simulations that were interesting but didn't end up being significant to the project. The flow chart and description of the beautifully structured program both died. The biggest cut came from editing the writing itself- text takes up a lot of room when it's double spaced.

As a team, how did you show that teamwork was involved in the project in your research report?

Tony: There are countless advantages to working on a team. First of all, the vast majority of all professional research is done by teams and groups of scientists rather than individuals. The reason for this? Most science projects, especially professional ones, are simply too massive for one person to undertake. Therefore, it is wise to break up a project into smaller, manageable parts. For example, our project had two clear, distinct sections. I was mostly in charge of putting together the model for objects in the upper atmosphere, while Erika implemented the code as well as made it efficient. When we gave our presentations, we presented the part of the project that we were most familiar with. However, each of us knew the other's work, and had an understanding of the project as a whole.

How much time did you put into this project?

Erika: Since we don't go to the same school, Tony and I meet for about two hours once a week. A lot of the work is individual. Probably on average I put in a five hour programming chunk a week, plus another hour or two for report writing. The work is also uneven throughout the year, deadlines make you get things done.

Tony: The majority of the work was done individually; the meetings were used to piece together our work and plan our next step ahead. For me, I spent anywhere from a few hours to 10+ hours a week depending on what I needed to do and when it needed to be done. For example, on the week that the final report was due, Erika and I spent around 8 hours on a single day writing and editing the report.

Do you think it's easier to work alone or as a team?

Tony: Personally, I believe that the advantages of working on a team outweighs the disadvantages. On a good, cooperative team, not only does more work get done, but the old saying "two heads are better than one" can really be seen. More people means more ideas.

Erika: I agree, teamwork is very advantageous. Once you learn how to work on a team, you stop having the 'elementary school project' type team where one person does everything. Teams are a great way to do bigger, more inter-disciplinary projects.

Is this what you want to study in college?

Erika: I think aerospace is a fascinating area, and I'd definitely like to study it more in the future. On the other hand, I also like physics, engineering, and computer science. Maybe I'll quadruple major?

Tony: As of now, I plan on majoring in Electrical Engineering; however, this is subject to change seeing that I also have interests in physics and aerospace.

What advice do you have for other kids entering this kind of competition?

Tony: The amount of time needed to research a project for a competition like Siemens is significant. My advice would be to start early, maybe even a year in advance. Find some topics that truly seem interesting, and see what recent developments have been made in those fields. Always be on the lookout for something unique, because that is what really catches people's eyes. Don't be afraid to change the path of the project. We had to change the direction of which way we wanted our project to take several times. Finally, ask around for people to help. There are many professional scientists who are willing to be a mentor figure for a student pursuing research. Just need to ask around!