Andrew Pastrello of Australia won the internship award at the Great Cross-Architecture Challenge. To demonstrate the simplicity of porting CUDA* code to Data Parallel C++ (DPC++), his application optimizes a music production tool to synthesize audio from gravitational waveforms produced by binary black hole inspiral-merger-ringdown simulations.

The Great Cross-Architecture Challenge was a 14-week contest for professional and student software developers who are interested in developing cross-architecture applications using oneAPI. Participants were challenged to be the next “oneAPI hero” by either porting an existing C/C++ or CUDA application using the Intel® DPC++ Compatibility Tool or by creating an entirely new oneAPI application. As part of the contest, developers received free access to Intel® toolkits on Intel® Developer Cloud across an array of Intel CPUs, GPUs and FPGA architectures, along with resources such as code samples, dev guides, webinars, and a developer collaboration portal (Intel® DevMesh).
 

Speaker

Andrew Pastrello is a PhD student at UNSW Sydney and a nuclear engineer at the Australian Nuclear Science and Technology Organisation (ANSTO). His research is in performance portable Monte Carlo neutron transport algorithms, and he is interested in programming new computer architectures.