Archived Content

The following content is from an older version of this website, and may not display correctly.

SoftLayer Technologies, a cloud infrastructure services provider and a recent IBM purchase, recently hosted infrastructure for a contest among developers of open-source robot-control software.

The Open Source Robotics Foundation, the contest's organizer, contracted with SoftLayer for bare-metal servers, cloud servers and high-performance GPU servers to support the DARPA Virtual Robotics Challenge. DARPA, or Defense Advanced Research Projects Agency, is an agency of the US Department of Defense that funded the simulator-based contest.

SoftLayer chief scientist Nathan Day said the competition, first of its kind, was an “exclamation point in the evolution of the cloud,” since it tested a cloud infrastructure's performance limits and requried cloud to be used together with bare-metal servers and virtual environments.

“SoftLayer’s platform can be uniquely tailored to meet requirements across the full spectrum of server needs and we are thrilled to work with Open Source Robotics Foundation in this premier event,” he added.

Competitors sought to develop software to enable a simulated robot to execute tasks similar to the activities that might be required of emergency personnel in a disaster response situation. The event drew more than 100 teams from around the world.

Winners of the virtual competition will move on to the next stage, in which their software will be applied to physical robots during a live event.

The foundation configured SoftLayer’s platform into a specialized format so that teams were able to control their own server constellations apart from other teams. Through SoftLayer’s API, each team was given five connected servers, including two high-end NVIDIA dual Intel Sandy Bridge servers with GPU, isolated from any others in the competition.

Teams were able to reload their own servers as needed, and OSRF could reset constellations to their virgin state once each team finished its simulation.