Open AccessResource-aware fluid scheduling with time constraints for clustered many-core architectures
Author(s) -
Anqi Yin,
Yi Guo,
Di Tang
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1971/1/012090
Subject(s) - computer science , scheduling (production processes) , integer programming , distributed computing , schedule , time complexity , parallel computing , resource consumption , execution time , job shop scheduling , preemption , mathematical optimization , linear programming , heuristic , algorithm , mathematics , ecology , artificial intelligence , biology , operating system
With the rise of many-core processors, real-time applications processing obtains better performance acceleration and parallelization freedom. A validate schedule is able to minimize system resource consumption under the time constraints, which has become a focus recently. This paper establishes an accurate fluid task model and an optimal integer linear programming (ILP) model further. These models not only realize the flexible parallel scheduling, but also achieve co-scheduling between computing and communication resources. Moreover, a heuristic algorithm is presented to generate near-optimal solutions within polynomial time. The experimental results show that the proposed approach is validate and capable to balance the time and resource consumption effectively. It leads the improvement of schedulability and validity up to 68.67% and 24.67% respectively, with 13.16% reduction of the system density averagely.