ETP4HPC webinar: "Task-Based Performance Portability in HPC"

11:00am to noon CET

As HPC hardware continues to evolve and diversify and workloads become more dynamic and complex, applications need to be expressed in a way that facilitates high performance across a range of hardware and situations. The main application code should be platform-independent, malleable and asynchronous with an open, clean, stable and dependable interface between the higher levels of the application, library or programming model and the kernels and software layers tuned for the machine. The platform-independent part should avoid direct references to specific resources and their availability, and instead provide the information needed to optimise behaviour.

The ETP4HPC white paper “Task-Based Performance Portability in HPC” summarises how task abstraction, which first appeared in the 1990s and is already mainstream in HPC, should be the basis for a composable and dynamic performance-portable interface. It outlines the innovations that are required in the programming model and runtime layers, and highlights the need for a greater degree of trust among application developers in the ability of the underlying software layers to extract full performance. These steps will help realise the vision for performance portability across current and future architectures and problems.

We have invited the authors of this White Paper to present their findings and recommendations. As usual, we will have a long Q&A session at the end to allow our audience to exchange with the speakers.

Agenda

• 11:00 Introduction and housekeeping information
• 11:05 Introduction to Principles of Task-Based Performance Portability: Illustration with the StarPU Runtime System for Heterogeneous HPC Platforms
  Olivier Aumage, INRIA
• 11:25 Task-Based Performance Portability with OmpSs
  Paul Carpenter, BSC
• 11:45 Q&A session

Speakers

Olivier Aumage is a full time researcher at Inria in Bordeaux since 2003; his main research topics include parallel runtime systems, programming models and languages for high performance computing applications.

Paul Carpenter is a Senior Researcher at Barcelona Supercomputing Center since 2011. His research interests include performance-portable parallel programming models, system software and energy-efficient architectures.