Data parallelism has arguably become the most important form of parallel programming. But according to parallel computing expert James Reinders, language support for it has been tucked away in single instruction, multiple data (SIMD) directives like OpenMP* or kernels like OpenCL™ standard or CUDA*.

The result? According to James, exploiting data parallelism is “too much in the weeds” and gets worse when you seek to be portable.

This is why he's excited about Data Parallel C++ (DPC++), a programming language and model that offers developers real advantages, including bringing heterogeneous programming to C++, code reuse from CPUs to GPUs to FPGAs, and cross-platform parallelism performance.

James discusses the developer benefits of DPC++, including:

  • The programming challenges it addresses
  • Key components of DPC++ and the standards it's based on
  • How DPC++ compares to other software development inflection points in history
  • A peek inside his new book focused on DPC++

Watch.

Download Data Parallel C++: Mastering DPC++ for Programming of Heterogeneous Systems Using C++ and SYCL, a book by James in its entirety. Additionally, you can download each of its 19 chapters as a stand-alone PDF. Get this free, in-depth reference that covers everything you need to get started with DPC++.


Other Resources

  • Explore this initiative led by Intel, including the download of free software toolkits like the essential Intel® oneAPI Base Toolkit, which includes the DPC++ compiler and libraries. Learn More
  • Sign up for an Intel® DevCloud account—a free development sandbox with access to the latest Intel® hardware and oneAPI software. No downloads. No configuration steps. No installations.

James Reinders
Engineer and author

James Reinders is an engineer who likes fast computers and the software tools to make them speedy. James has spent more than 30 years in high-performance computing (HPC) and parallel computing, including 27 years (10,001 days to be precise) at Intel Corporation. His list of industry contributions includes the world's first TeraFLOP/s supercomputer (ASCI Red), the world's first TeraFLOP/s microprocessor (Intel® Xeon Phi™ processor), and Threading Building Blocks (TBB). Additionally, he coined the phrases "Parallel or Perish" and "Think Parallel". James is an author of eight technical books (including the original book on TBB), numerous papers, articles, blogs, and has co-taught many technical classes, workshops, and sessions. James earned a master of science in electrical engineering from the University of Michigan, Go Blue.

Lindsay Michelet
AI marketing and messaging strategist, Intel Corporation

Lindsay Michelet comes with a passion for honest storytelling about the real-world impact of Intel® products and solutions. Her background includes more than 16 years of integrated business-to-business (B2B) marketing communication strategy and account direction with expertise in building messaging and content platforms, partner marketing programs, and go-to-market campaigns for various industries and initiatives including AI, IoT, and autonomous driving. As part of the AI Marketing team, Lindsay oversees cross-Intel AI messaging strategy and implementation. She holds a bachelor's of arts in marketing from Portland State University in Oregon.

 

Intel® oneAPI Base Toolkit

Get started with this core set of tools and libraries for developing high-performance, data-centric applications across diverse architectures.

Get It Now   

See All Tools