Programs
- M. Tech. in Automotive Engineering -
- Fellowship In Interventional Pulmonology and Pulmonary Critical care - Fellowship
Publication Type : Journal Article
Source : Real-Time Syst, (2022)
Url : https://link.springer.com/article/10.1007/s11241-022-09389-4
Campus : Coimbatore
School : School of Engineering
Department : Computer Science and Engineering
Verified : No
Year : 2022
Abstract : In this paper, we propose a new model for mixed-criticality systems (MCS) that can adapt to system overload scenarios due to the occurrence of two known phenomena. First, budget overrun of tasks due to their imprecise budget estimations which is a common theme in most existing studies related to MCS. Second, a spike in the resource consumption pattern leads to a temporary system overload when tasks belonging to different modes execute together for a certain time interval when the system switches from one mode to another. Such a phenomenon is also a common theme for designing safety-critical embedded systems capable of operating in multiple functional modes. The multi-mode mixed-criticality system (MM-MCS) model, proposed in this paper, considers task degradation and offsets to handle the system overload. We derive a sufficient t schedulability test under a fixed-priority scheduling scheme for the MM-MCS model and present an algorithm to determine offset values for tasks with an objective to minimize the offsets for higher criticality tasks. Experimental results with synthetic task sets show that task degradation and offsets are an effective means to handle system overload both within a mode and across mode transitions. Further, an algorithm to derive offsets for tasks based on their criticality is presented. It clearly shows that the criticality of tasks can play an important role not only in deciding on the task degradation but also in determining offset values for tasks.
Cite this Research Publication : Sundar, V.K., Ramanathan, S. & Easwaran, A. "Design and analyses of functional mode changes for mixed-criticality systems". Real-Time Syst (2022). https://doi.org/10.1007/s11241-022-09389-4