oneAPI

Code Together is an interview series that explores challenges at the forefront of cross-architecture development, sponsored by Intel. We’ll talk with those across the industry who are forging a path on this often treacherous journey through an increasingly diverse, data-centric world. Join the conversation.

Paul Navrátil
Director of Visualization, Texas Advanced Computing Center (TACC)
Jim Jeffers
Sr. Principal Engineer and Sr. Director of Advanced Rendering and Visualization, Intel

We explore the critical role of visualization and visual analytics in disaster management and medical research with Paul Navrátil, Director of Visualization at the Texas Advanced Computing Center (TACC), and Jim Jeffers, Sr. Principal Engineer and Sr. Director of Advanced Rendering and Visualization at Intel. Paul reveals TACC’s approach to recent weather crises, as well as their partnership with National Oceanic and Atmospheric Administration (NOAA) in hurricane forecasting. He also talks about their collaboration with the Open Science community in the national HPC mobilization effort to fight COVID-19. Jim unpacks the complexities of in-situ analysis, ray tracing and line-of-sight physics. And they jam about the SOLAR Ray Tracing Consortium, Khronos Group ANARI Working Group and more.

To learn more:

Intel oneAPI Rendering Toolkit

https://www.oneapi.com/

www.tacc.utexas.edu/

Stampede II

Frontera

SOLAR Ray Tracing Consortium

WARRAS.SOLARRT.org

Khronos ANARI Working Group

Open Science community

Andrey Alekseenko
Postdoctoral Fellow, KTH Royal Institute of Technology & SciLifeLab
Erik Lindahl
Biophysics Professor, Stockholm University & KTH Royal Institute of Technology
Roland Schulz
Parallel Software Engineer, Intel

In this second episode about GROMACS, one of the world’s most widely used open source molecular dynamics (MD) applications, we explore the quest to simplify portable, performant programming with Erik Lindahl, Biophysics Professor at Stockholm University & KTH Royal Institute of Technology, and Roland Schulz, Parallel Software Engineer at Intel. Andrey Alekseenko, Postdoctoral Fellow at KTH Royal Institute of Technology and SciLifeLab, shares his experiences porting GROMACS across hardware architectures. And our guests talk about Khronos SYCL’s rapid growth and vibrant, thriving community.

To learn more:

http://www.gromacs.org/

https://gitlab.com/gromacs/gromacs

https://github.com/topics/sycl

https://github.com/intel/llvm

https://www.oneapi.com/

Jeff Hammond
Jeff Hammond
Principal Engineer, Intel

The oneAPI specification allows you to build standards-driven code, with a quicker path to performance, across CPUs, GPUs, FPGAs and other accelerators. It supports cross-architecture programming through a language, set of library APIs and a low-level hardware interface, as well as a reference stack. The specifications and source code are built on GitHub and integrate with upstream and downstream workflows. Intel Principal Engineer Jeff Hammond demonstrates how oneAPI works on a range of hardware and is built using GitHub. Contribute to the specification or the open source implementation. Use the base oneAPI software stack and port it to your favorite platform.

Watch Now Download Presentation

Aksel Alpay
Jeff Hammond
Aksel Alpay
Engineer, Heidelberg University Computing Center
Jeff Hammond
Principal Engineer, Intel

Open standards underpin the future of heterogeneous programming, and as Intel Principal Engineer Jeff Hammond points out, SYCL has proven in many ways that “standards don’t cost you a thing and you get portability.” A modern C++ language, SYCL has demonstrated amazing growth and now boasts four implementations—ComputeCpp, triSYCL, hipSYCL and DPC++. hipSYCL Creator Aksel Alpay explains some of the differences between these implementations, and how hipSYCL is evolving to incorporate DPC++ features contributed to the SYCL 2020 specification, including unified shared memory (USM), reductions, group algorithms and subgroups. The two also dive into a discussion of buffers and accessors, and Aksel debunks a few myths around hipSYCL.

To learn more:

hipSYCL (GitHub)

Dive into DPC++ (Blog)

DPC++ Data Management across Multiple Architectures (Blog)

Data Parallel C++ Book

oneAPI

Penporn Koanantakool
AG Ramesh
Penporn Koanantakool
Senior Software Engineer, Google
AG Ramesh
Principal Engineer, Intel

Increased processing power, massive amounts of data, and the development of more advanced algorithms have brought deep learning to the forefront, and TensorFlow has emerged as one of the world’s most popular machine learning frameworks. Penporn Koanantakool, Google Senior Software Engineer, and Ramesh, Intel Principal Engineer, share how their teams are collaborating to optimize TensorFlow for the latest Intel technologies using oneAPI Deep Neural Network Library (oneDNN). The result: remarkable performance gains that will benefit applications spanning natural language processing, image and object recognition, autonomous vehicles, fraud detection, medical diagnosis and treatment, and much more. Intel-optimized TensorFlow is now made available through Intel® AI Analytics Toolkit and is being used within Google Cloud Platform and a Google Health project.

To learn more:

Intel® oneAPI Deep Neural Network Library

Intel® AI Analytics Toolkit

Accelerating DeepVariant with Intel’s AVX-512 Optimizations

TensorFlow-MKL int8 Optimizations for Cascade Lake

TensorFlow-MKL bfloat16 Optimizations for Cooper Lake

Andrew Lumsdaine
Chief Scientist, Northwest Institute for Advanced Computing
Mike Voss
Principal Engineer, Intel

Andrew Lumsdaine, Chief Scientist at Northwest Institute for Advanced Computing, shares his journey in parallelism, from the HPC community’s early skepticism of and eventual move to C++ to build large systems, to his pursuit to develop a series of C++ libraries to strike at the contradiction between abstraction and performance, to today’s reality that parallelism and heterogeneity are essential for achieving performance. Together, Andrew and Mike Voss, Principal Engineer at Intel, underscore the value of open standards and high-quality implementations of open standards—specifically, the oneAPI specification and Intel® oneAPI toolkits, respectively—and provide listeners with ways to get involved in helping shape ISO C++, Khronos SYCL, DPC++ and oneAPI.

To learn more:

×

Learn DPC++ for Free

Learn More