Associate Professor of Physics, Applied Physics, and Astronomy Heidi Newberg has been working with the Sloan Digital Sky Survey, a project that involves scanning segments of the Milky Way Galaxy. Though Newberg had this data, she wanted to build a digital structure of the galaxy.
According to Newberg, the importance of building a model of the Milky Way is to discover more of its history and learn more about dark matter—particles that are theoretically comprised of a major portion of mass in the galaxy and influence orbitary movements of space material.
At first, Newberg tried to create a model by setting fixed parameters to data from the Sloan project. However, she was unsuccessful in using just one computer, as there were too many variables to take into account. “If you have a problem that is really critical and you have the money, you can build a super computer to solve it for you. [But] no one was going to build me a super computer,” she joked. That’s when she went to the computer science department for help.
As part of his thesis, now Post Doctoral Research in Computer Science Travis Dessell created MilkyWay@home with the help of The Berkeley Open Infrastructure for Network Computing, or BOINC for short. After receiving the astronomy code from Newberg, Dessell worked on making it a publicly available application to help produce the Milky Way model.
Conceived about two-and-a-half years ago, the project is the first volunteer computing project for RPI. In that time alone, MilkyWay@home has become the third fastest computing system in the world, surpassing even SETI@home (a project whos goal is to find intelligent life).
One thing that Dessell noted to MilkyWay@home’s success was the openness of the project. “People can join and freely look at the source code—they like the communication,” he commented. Doing this allows people to look at the code and even make it more efficient should they spot ways to improve the computing system.
Dessell also encouraged usage of MilkyWay@home by RPI students, noting that it was a good opportunity for undergraduates to either passively or actively be involved with research at the Institute. Simply running the program enables the user to partake in a bigger piece of research, while those who are more interested can come and join the project as part of undergraduate research, something that Dessell welcomes. The project is unique in that it can be tailored to a student’s interest, either in astronomy or computer science.
Another reason to help out is to simply be a part of something RPI-related. “it gives a sense of school pride,” stated Dessell when talking about the success of the project.
Dessell went on to talk about the future production of another application, DNA@home, which will look into protein transcription sites on DNA. The project, researched by Lee Newberg, Newberg’s husband, will hope to see its alpha release in the coming weeks and will be the second volunteer computing project to come to RPI.
For more information about MilkyWay@home and to find out how you can help, visit http://milkyway.cs.rpi.edu/milkyway/.