Intel has announced the winners of the Great Cross-Architecture Challenge, a collaboration with the European Organization for Nuclear Research (CERN) and the Argonne National Laboratory led by CodeProject.
Summer CERN Openlab Internship Winner or $ 8,000 Cash:
- Eugenio Marinelli of Sophia Antipolis, France, used oneAPI’s cross-architecture libraries and tools to develop a new application that decodes digital data stored in synthetic DNA. This new storage method offers simple, fast and inexpensive storage of data archives that can take centuries even in harsh environments.
Winner of a chance to work on a oneAPI project with Argonne or $ 8,000 in cash:
- Andrew Pastrello from Sydney, Australia showed how easy it is to port CUDA code and modified the application using DPC ++ and oneAPI tools to synthesize audio from gravitational waveforms generated by black hole simulations for music production and the Astronomy lessons were generated.
Trip Winner to CERN or $ 5,000 Cash:
- Rafael Campos from Lisbon, Portugal demonstrated the development of oneAPI by customizing OpenCL applications to improve the performance of bioinformatic applications. The solution has the potential to improve epistasis detection and detect diseases such as Alzheimer’s and various cancers at an early stage.
- Zhen Ju from Shenzhen, China, demonstrated the easy migration of a CUDA-based application and the advantages of an open programming model that serves different architectures. The application port filters redundant sequences in genetic research and benefits pharmaceutical discovery, agriculture, and medical care.
- Ricardo Nobre of Lisbon, Portugal used the Intel DPC ++ Compatibility Tool to port a CUDA-based application, with more than 95 percent of the code migrated automatically. The application uses CPU and GPU devices to discover new associations between genotypes and phenotypes, potentially leading to improved screening, personalized treatments and drug development.
The challenge challenged software developers to use oneAPI to create fast, efficient and future-proof applications that use various XPUs, including CPUs, GPUs, FPGAs and other accelerators. Attendees had access to free resources such as code samples, developer guides, webinars, and the DevMesh portal for developer collaboration to accelerate their development.
The five winners were selected by a jury of six judges who rated the competition entries on the basis of innovation, impact on humanity, use of cross-architecture computing, general benefit to the target audience, level of coding skills and quality of the project explanation. Twenty applicants will receive a cash prize of $ 500 for their submissions. The challenge drew participants on five continents. Participants used oneAPI and Data Parallel C ++ (DPC ++) to build applications in areas such as bioinformatics, cryptography, data analysis, education, financial services, genomics, healthcare, image processing, math, molecular dynamics, and particle physics, and ray tracing.
Attendees used free Intel oneAPI toolkits and the Intel DevCloud, which can be used to test code and workloads in a variety of Intel XPU architectures to upgrade an existing C / C ++ application, a CUDA (Compute Unified Device Architecture) to DPC ++ or to create a completely new oneAPI application for working with multiple architectures.
“The participants in the Great Cross-Architecture Challenge demonstrated the potential of oneAPI,” said Maria Girone, Chief Technology Officer of CERN openlab. “Using them, they were able to write code for heterogeneous hardware architectures with a wide variety of applications. People from all over the world were able to access cutting edge technology through this developer challenge. We look forward to welcoming the winners of the competition at CERN. “
“The challenge was an exciting opportunity to explore oneAPI and its ability to develop applications for heterogeneous computing architectures,” said Katherine Riley, director of science at Argonne Leadership Computing Facility. “The contributions covered a variety of topics and used several approaches that illustrated the enormous potential of oneAPI – and the creativity of the participants! We look forward to bringing this creativity to Argonne as we prepare for our upcoming Aurora exascale system. “