Resource Efficient Isolation Mechanisms in Mixed-Criticality Scheduling

Abstract

Mixed-criticality real-time scheduling has been developed to improve resource utilization while guaranteeing safe execution of critical applications. These studies use optimistic resource reservation for all the applications to improve utilization, but prioritize critical applications when the reservations become insufficient at runtime. Many of them however share an impractical assumption that all the critical applications will simultaneously demand additional resources. As a consequence, they under-utilize resources by penalizing all the low-criticality applications. In this paper we overcome this shortcoming using a novel mechanism that comprises a parameter to model the expected number of critical applications simultaneously demanding more resources, and an execution strategy based on the parameter to improve resource utilization. Since most mixedcriticality systems in practice are component-based, we design our mechanism such that the component boundaries provide the isolation necessary to support the execution of low-criticality applications, and at the same time protect the critical ones. We also develop schedulability tests for the proposed mechanism under both a flat as well as a hierarchical scheduling framework. Finally, through simulations, we compare the performance of the proposed approach with existing studies in terms of schedulability and the capability to support low-criticality applications.

Towards compositional mixed-criticality real-time scheduling in open systems

“…Using component-mode, they identify fine-grained execution behaviors of jobs in a component. However, they still have pessimism that all components request up to their HI-criticality budgets when a single component requests its HI-criticality budget [7] or all LO-tasks in a component are dropped at component mode-switch [9]. Our task-mode identifies more fine-grained execution behaviors in the system.…”

“…We schedule 1 RMD is the reverse of deadline miss ratio (DMR): RM D = 1 − DM R 2 If the system mode is HI (LO), current jobs of all tasks may execute up to their HI-WCETs (LO-WCETs). 3 Its example can be found in Gu et al [7]. a LO-task with its real deadline.…”

“…Overcoming the limitation of system mode, compositional MC approaches [7,9] introduced a concept of component-mode, which represents runtime criticality mode for a component. Using component-mode, they identify fine-grained execution behaviors of jobs in a component.…”

“…Little attention has made to efficiently scheduling mixed-criticality components in a compositional manner in open systems, except recent studies [7,9]. Although traditional temporal isolation only consider strict isolation between components, we plan to consider a diverse levels (degrees) of temporal isolation in consideration of mixedcriticality.…”

“…However, such an assumption is very unlikely to happen in practice. Recently, a few studies considered this issue [5,7,9,11] and proposed approaches to improve resource utilization for low-criticality tasks. We plan to consider this issue for compositional mixed-criticality scheduling in open systems.…”