The Texas Advanced Computing Center (TACC) at The University of Texas at Austin reports it is using the Intel oneAPI Rendering Toolkit for remote analysis capabilities on the center’s supercomputers, “providing high-fidelity visualization tools for all scales of computing.”
TACC, which operates the Frontera and Stampede2 supercomputer along with more than a dozen HPC systems, said the adoption of oneAPI is part of an effort to advance its compute and visualization libraries, such as Galaxy. TACC said it will use oneAPI, Intel ray tracing libraries and Intel CPU and Xe GPU hardware for scientific simulations for in situ visual analysis.
oneAPI is a cross-architecture programming model for CPUs and accelerator architectures (GPUs, FPGAs, and others) designed to simplify software development and enable accelerated compute without proprietary lock-in and support integration of legacy code. “With oneAPI, developers can choose the best architecture for the specific problem they are trying to solve without needing to rewrite software for the next architecture and platform,” Intel said.
The oneAPI Rendering Toolkit offers open source libraries for creating visualization applications. The libraries have rendering kernels and middleware for Intel platforms and supports Intel CPUs and future Xe architectures (GPUs). It includes the Intel Embree, Intel Open, Image Denoise, Intel, Open Volume Kernel Library, Intel OSPRay, Intel OpenSWR, and other components and utilities.
“Flexible, unified programming and optimized performance through Intel oneAPI Toolkits will enable high-fidelity, interactive analysis across TACC computing platforms without having to implement device-specific routines for each system,” reported TACC, which is an Intel Graphics and Visualization Institute of Xellence (Intel GVI).
Kitware‘s ParaView and Intelligent Light‘s VisIt are two TACC-supported applications enabled by the oneAPI Rendering Toolkit. “These compute and visualization capabilities provide users around the world the ability to perform high-fidelity analysis on data generated on TACC systems,” the center said, “without having to migrate to a separate machine, and this flexibility empowers TACC system designers to focus on maximizing full-system capabilities without incorporating a separate analysis subsystem.”
“Paul and the teams at TACC and Kitware are foundational to the Intel GVI’s,” said Jim Jeffers, senior principal engineer and senior director of Intel Advanced Rendering and Visualization. “They develop advanced analysis codes like Galaxy, the Visualization ToolKit, and perform large-scale testing of widely-used visualization applications such as ParaView and VisIt. Particularly important are the computation resources and support for all Intel GVIs that TACC provides.”
“TACC’s expanded role as a oneAPI Center of Excellence will be a cornerstone enabling top scientists globally to tackle climate change, sustainable energy generation, pandemics, and many more pressing challenges.”