This document summarizes research on self-tuning schedulers for legacy real-time applications. It describes a proposed approach using application monitoring to infer scheduling parameters like execution time and period, which are then used by the operating system to automatically provision real-time guarantees without source code modifications. Experimental results show this approach improves performance for legacy multimedia applications compared to general-purpose operating systems or real-time extensions. Future work is discussed to better support multi-threaded applications and multi-core systems.
Self-tuning Schedulers for Legacy Real-Time Applications
1. RETIS Lab
Real-Time Systems Laboratory
“Self-tuning Schedulers for Legacy Real-
Time Applications”
Tommaso Cucinotta, Fabio Checconi Luca Abeni, Luigi Palopoli
Real-Time Systems Laboratory University of Trento
Scuola Superiore Sant'Anna