VIDEO: NASA leans on CUDA for flight plan optimisation

A parallel computing platform and programming model called CUDA is being used to optimise flight planning for air traffic controllers, making airline travel safer and with fewer delays and congestion.
CUDA, invented by NVIDIA, enables dramatic increases in computing performance by harnessing the power of the graphics processing unit (GPU).
NVIDIA reckons the baseline air traffic control software, developed by NASA, currently takes about 10 minutes to perform a 24-hour trajectory prediction in four dimensions (latitude, longitude, altitude and time), assuming 35,000 aircraft will be flying on any given day.
With the CUDA prototype version, the application took less than 2.5 seconds, a factor increase of 250.
The CUDA port of the trajectory software is being developed by NASA and Optimal Synthesis, a Silicon Valley research firm that specialises in various types of software for engineers and scientists.
According to Monish Tandale, a research scientist at Optimal Synthesis, being able to compute trajectories with CUDA technology has a number of advantages.
Tandale explains how GPU computing exploits parallelism in the trajectory prediction process to have extremely fast run-times.
This in turn allows a leap in real-time performance and allows model analysis of a much greater complexity. It also opens up the possibility of using algorithms and approaches that were earlier deemed impractical due to the computational complexity.
Monish maintains that porting the code to GPUs is relatively easy thanks to the shallow learning curve of CUDA, allowing them to smoothly transition from C/C++ based CPU programming to GPU programming.

External Links

Posted in Innovation, News

Comments are closed.