Accelerated computing is predominantly done through a proprietary approach, but there is an open, community-driven alternative. The oneAPI industry initiative offers a platform that supports processor architectures from multiple vendors. This panel discusses the strengths of oneAPI, as well as gaps that must be addressed to successfully drive this open, standard approach into the accelerated computing community.

Speakers

James Reinders is an engineer at Intel, and is an author, coauthor, and editor of ten technical books related to parallel programming. His latest book is about SYCL*, and is available through a free download: Data Parallel C++: Mastering DPC++ for Programming of Heterogeneous Systems Using C++ and SYCL. James's parallel computing experience spans four decades, and he is focused on helping enable parallel programming in a heterogeneous world.

Maria Soledad Elli joined Intel in 2017 as data scientist focusing on automated vehicle (AV) safety analysis using simulation and AV safety standards and regulations. As part of the Corporate Strategy and Ventures team, Maria helps the team derive insights from different applications to advance Intel’s future and growth. In 2013, she received her bachelor of science degree in computer engineering at the National University of Tucuman, Argentina. She worked until 2015 as a software engineer for radar applications at the Aerospace and Government Division at INVAP SE, one of the leading Latin American corporations in applied high technology.

Felix LeClair is an open source engineer focused on high-performance computing (HPC), Single Instruction Multiple Data (SIMD), and SYCL. He specializes in optimizing software performance and enabling developers to make the most of cutting-edge hardware technologies that include CPUs, GPUs, or FPGAs.

Istvan Zoltan Reguly got his master of science in 2010 and his PhD in 2014 from Pázmány Péter Catholic University Information Technology (PPCU ITK) in computer science. He leads the high performance computing lab at PPCU ITK, where they do research into the design and implementation of domain specific languages for high performance computing.

Zhuldyz-Zhan Sagimbayev works at Cerebra as a lead machine learning engineer. He has been working with machine learning for about five years. He has a master’s degree from Kazakh-British Technical University, which is a top technical university in Kazakhstan.

Dr. Jian Huang is a professor of computer science in the Electrical Engineering and Computer Science (EECS) department at the University of Tennessee (UT). He started as an assistant professor at UT after graduating with his PhD in computer science from Ohio State University in 2001. His research is in data visualization, analytics, and HPC. His work on Visualization as a Service (VaaS) has expanded use case scenarios of data-intensive visualizations for widespread accessibility, shareability, reproducibility and replicability. His work is funded by the National Science Foundation (NSF), Department of Energy, Department of Interior, and Intel.

Hartwig Anzt is the director of the Innovative Computing Laboratory (ICL) and professor in the EECS department of the UT. He also holds a senior research scientist position at Steinbuch Centre for Computing at the Karlsruhe Institute of Technology. Hartwig holds a PhD in applied mathematics and specializes in iterative methods and preconditioning techniques for next-generation hardware architectures. He is author of the Matrix Algebra on GPU and Multi-core Architectures (MAGMA) Sparse open source software package and is the managing lead of the Ginkgo math software library. Hartwig is the principal investigator (PI) of software technology projects that are part of the US Exascale Computing Project (ECP), including a coordinated effort aimed at integrating low-precision functionality into high-accuracy simulation codes. He is also a PI in the EuroHPC Project MICROCARD.

Eric Nielsen is a senior research scientist with the Computational AeroSciences branch at the NASA* Langley Research Center in Hampton, Virginia. He received his PhD in aerospace engineering from Virginia Tech and has worked at Langley for 30 years.

Mohammad Zubair is a professor of computer science at Old Dominion University. His primary interest is performance and portability issues on high-performance emerging architectures for scientific computing and big data analytics. Mohammad collaborates with NASA Langley Research Center, Intel, AMD*, and Fermilab in porting and optimizing large scientific codes on emerging high-performance architectures. In the past, he has worked as a research staff member at the IBM* Thomas J. Watson Research Center, where he focused on developing optimized implementations of scientific kernels.