Optimize Task-based Programming in a Cross-architecture World

Adding parallelism to applications hasn’t been a nice-to-have for a while. If you want your apps to run on today’s hardware, you’ve got to parallelize it at the very least.

One way to simplify the process is by using Intel® oneAPI Threading Building Blocks (oneTBB), a task-based programming library for both mainstream and computationally complex applications that gives developers the right level of abstraction to build algorithms that perform optimally across heterogeneous systems.

Even if you’re not a threading expert.

Join Technical Consulting Engineer James Tullos for a tour of the latest oneTBB release. Topics covered include:

  • An overview of the library, including how it uses tasks or the oneTBB flow graph API to parallelize your application
  • How to use oneTBB in a cross-architecture, data parallel application that uses CPUs and GPUs
  • A demo of the library in a heterogeneous environment
  • The scalability exhibited by a flow graph

Download the software
Get the oneTBB as part of the Intel® oneAPI Base Toolkit, a core set of tools and libraries for developing performant applications across diverse architectures.

Resources

  • Sign up for an Intel® DevCloud for oneAPI account—a free development sandbox with access to the latest Intel® hardware and oneAPI software.
  • Explore oneAPI, including developer opportunities and benefits
  • Subscribe to the POD—Code Together is an interview series that explores the challenges at the forefront of cross-architecture development. Each bi-weekly episode features industry VIPs who are blazing new trails through today’s data-centric world. Available wherever you get your podcasts.
James Tullos, Technical Consulting Engineer, Intel Corporation

James Tullos is a Technical Consulting Engineer responsible for all things “customer technical support and training” on application performance using Intel Intel® Software Development Tools. He focuses on parallel performance in HPC and cluster environments, with specific knowledge around distributed memory computing using the Message Passing Interface (MPI).

James enjoys sharing his knowledge and is a regular presence at technical trade shows, workshops, and conferences. Prior to joining Intel, he worked in the aerospace engineering field focused on propulsion system analysis programs.

James holds an MS in Aeronautical and Astronautical Engineering from Purdue University, and a BS in Aerospace Engineering from Mississippi State University.

Performance varies by use, configuration, and other factors. Learn more at www.Intel.com/PerformanceIndex.