Title :
LP based MPC Algorithm in Distributed Real-Time Systems with End-to-End Tasks
Author :
Fangling, Zhang ; Jinbiao, Wang
Author_Institution :
Dept. of Autom., Tsinghua Univ., Beijing
Abstract :
This paper presents a linear-programming (LP) based model predictive control (MPC) algorithm to handle the utilization control in distributed real-time systems with end-to-end tasks. The LP based MPC algorithm can achieve desired CPU utilization despite some uncertainties of the system workload. And the computational advantage of LP problems makes it promising to handle large-scale systems. The simulations demonstrate that the algorithm exhibits good robustness to task execution times uncertainty while having little computational cost
Keywords :
distributed control; large-scale systems; linear programming; predictive control; real-time systems; distributed real-time system; end-to-end task execution time; large-scale system; linear-programming based model predictive control algorithm; utilization control; Computational efficiency; Computational modeling; Distributed control; Large-scale systems; Prediction algorithms; Predictive control; Predictive models; Real time systems; Robustness; Uncertainty;
Conference_Titel :
Computational Intelligence for Modelling, Control and Automation, 2005 and International Conference on Intelligent Agents, Web Technologies and Internet Commerce, International Conference on
Conference_Location :
Vienna
Print_ISBN :
0-7695-2504-0
DOI :
10.1109/CIMCA.2005.1631608
