Offload Your Code from CPU to GPU and Optimize It

Locating and removing bottlenecks is an inherent challenge for every application developer. It’s made more complex when porting an application to a new platform, for example, from a CPU to a GPU. Developers who do that not only must identify bottlenecks, they must figure out which parts of the code benefits from offloading in the first place.

In this webinar, software optimization expert Kevin O’Leary discusses how Intel® Advisor helps developers remove these new CPU-to-GPU porting obstacles.

He covers:

• Offload Advisor: A command-line feature that projects performance speedup on accelerators and estimates offload overhead
• GPU Roofline Analysis: A technical preview that identifies bottlenecks in GPU-ported code and shows how close its performance is to system maximums
• A Demonstration: Get a walk-through of a matrix multiplication example to learn how the previous features can help you optimize application efficiency for GPUs

Technical Level: Intermediate

• Download Intel Advisor as part of the Intel® oneAPI Base Toolkit, a core set of tools and libraries for creating data-centric, cross-architecture applications.
• Sign up for an Intel® Developer Cloud account—a free development sandbox with access to the latest Intel hardware and oneAPI software.

• Offload Cookbook: Identify Code Regions to Offload to GPUs

Kevin O'Leary

Senior software developer and lead technical consulting engineer, Intel Corporation

Kevin O’Leary's expertise includes compilers, debuggers, and software performance tools. He’s currently responsible for performance optimization using Intel Advisor and Intel® VTune™ Profiler and was a key developer of the Intel® Parallel Studio XE development suite. Before joining Intel, Kevin spent many years as a debugger engineer for IBM* and Rational Software*.

Kevin holds a BA degree in computer science from the University of Massachusetts and an MA degree in computer science from Oregon Health and Science University.