LONDON Graphics chipmaker Nvidia Corp. (Santa Clara, Calif.) has said it plans to offer prizes totaling $5,000 to five programmers who use the Cuda parallel programming environment to solve the first of a series of challenges.
The so-called Cuda Superhero challenges are being posted at
www.topcoder.com/nvidia, the company said.
The first contest begins Sept. 14 and will conclude Sept. 25 and the five winners will be announced at Nvidia's GPU technology conference being held at the Fairmont Hotel in San Jose, Calif., from Sept. 30 to Oct. 2, 2009.
Other contests with prizes are expected to follow. The top prize for the first event is $2,500 but the specification of the problem and how the competitors would be scored was not stated.
"PC architecture has evolved from central processing on just the CPU to co-processing on the CPU and the GPU," said Sanford Russell, general manager of the Cuda group at Nvidia. "By tapping into the TopCoder community, we can educate over 200,000 programmers on the advantages parallel programming offers and fan the flames of the GPU computing revolution."
GPU computing uses the massively parallel architecture of the graphics processing unit (GPU) as a computational engine using high-level languages and APIs, such as C and Fortran, or programming interfaces, such as DirectCompute and OpenCL. The model for GPU computing is to use a CPU and GPU together in a co-processing computing model.
"We anticipate that the CUDA Superhero Challenge will generate many creative approaches to solving computing challenges and will result in game-changing innovations," said Rob Hughes, president and chief operating officer at TopCoder.
Related links and articles:
Researchers report progress on parallel path
Parallel software plays catch-up with multicore
Unlocking the promise of graphics processors
GP GPU: th evolution of the coprocessor