Exploring scheduling algorithms for parallel task graphs: a modern game engine case study
Résumé
Game engines are at the heart of the design of modern video games. One of their functions is to keep a high frame rate by scheduling the tasks required to generate each frame (image). These tasks are organized in a soft real-time, parallel task graph, which is a scenario very few works have focused on, or adapted scheduling algorithms to. In this paper, we study the scheduling problem of game engines. We model the tasks and the scheduling problem by proling a commercial game engine, adapt and compare dierent scheduling algorithms, and propose two additional scheduling optimizations.
Origine : Fichiers produits par l'(les) auteur(s)