Overall Objectives
Research Program
Application Domains
New Software and Platforms
New Results
Bilateral Contracts and Grants with Industry
Partnerships and Cooperations
XML PDF e-pub
PDF e-Pub

Section: New Results

Modeling and Simulating of Dynamic Task-Based Runtime Systems

Participants : Luka Stanisic [PhD, Inria, Mescal] , Samuel Thibault [Univ. Bordeaux, Inria, Storm] , Brice Videau, Arnaud Legrand [CNRS, Inria, Mescal] , Jean François Méhaut.

Multi-core architectures comprising several GPUs have become mainstream in the field of High-Performance Computing. However, obtaining the maximum performance of such heterogeneous machines is challenging as it requires to carefully offload computations and manage data movements between the different processing units. The most promising and successful approaches so far build on task-based runtimes that abstract the machine and rely on opportunistic scheduling algorithms. As a consequence, the problem gets shifted to choosing the task granularity, task graph structure, and optimizing the scheduling strategies. Trying different combinations of these different alternatives is also itself a challenge. Indeed, getting accurate measurements requires reserving the target system for the whole duration of experiments. Furthermore, observations are limited to the few available systems at hand and may be difficult to generalize. In this work, we show how we crafted a coarse-grain hybrid simulation/emulation of StarPU, a dynamic runtime for hybrid architectures, over SimGrid, a versatile simulator for distributed systems. This approach allows to obtain performance predictions of classical dense linear algebra kernels accurate within a few percents and in a matter of seconds, which allows both runtime and application designers to quickly decide which optimization to enable or whether it is worth investing in higher-end GPUs or not. Additionally, it allows to conduct robust and extensive scheduling studies in a controlled environment whose characteristics are very close to real platforms while having reproducible behavior.

This work is part of the Luka Stanisic's thesis. Luka stanisic was coadvised by Arnaud Legrand, Brice Videau and Jean-François Méhaut. This thesis was defended in November 2015. Luka Stanisic currently holds a postdoc position at Inria Bordeaux in the Storm and HiePacs teams. This work was published in the CCPE journal [9] .