Finite ready queues as a mean for overload reduction in weakly-hard real-time systems

Abstract

Finite ready queues, implemented by bu ers, are a system reality in embedded real-time computing systems and networks. The dimensioning of queues is subject to constraints in industrial practice, and often the queue capacity is su cient for typical system behavior but is not su cient in peak overload conditions. This may lead to over ow and consequently to the discarding of jobs. In this paper, we explore whether nite queue capacity can also be used as a mean of design in order to reduce workload peaks and thus shorten a transient overload phase. We present an analysis method which is to the best of our knowledge the rst one able to give (a) worst-case response times guarantees as well as (b) weakly-hard guarantees for tasks which are executed on a computing system with nite queues. Experimental results show that nite queue capacity may only a have weak overload limiting e ect. This unexpected outcome can be explained by the system behavior in the worst-case corner cases. The analysis shows nevertheless that a trade-o between weakly-hard guarantees and queue sizes is possible.

Generalized Weakly Hard Schedulability Analysis for Real-Time Periodic Tasks

“…A control design which explicitly takes into account missed deadlines in a probabilistic fashion is recently explored in [28]. Several recent papers like [1,2,34] elaborate the use of weakly hard model in specific applications, including energy variable systems, finite queue platforms, cost-aware scheduling and networked environments, showing a constantly growing interest in the topic.…”

A Generic Coq Proof of Typical Worst-Case Analysis

“…The analysis of weakly-hard real-time systems is currently the subject of renewed attention in the research community [5], [6], [30] as well as in industry [15], [18], [26], [33].…”

“…Typical Worst-Case Analysis (TWCA) is one exception which can handle tasks with complex activation This work has been partially supported by the LabEx PERSYVAL-Lab (grant ANR-11-LABX-0025-01) through the CASERM project and the French national research organization ANR (grants ANR-15-CE25-0008 and ANR-17-CE25-0016) through the VOCAL and RT-PROOFS projects. models [27], task dependencies [18], finite ready queues [6] and even recently multiprocessor architectures [5]. Still, the approach needs improving.…”

“…Note that the approach we use here to bound the number of deadline misses can be easily adapted to count other events such as properties on task chains or generalized deadline misses [6].…”

“…Typical Worst-Case Analysis (TWCA) [27] is another approach which can handle not only tasks with complex activation models [27], but also task dependencies [18], finite ready queues [6] and, even recently, multiprocessor architectures [5]. Work in this area progresses in two directions: generalization to other scheduling policies and more realistic system models on the one hand, and reduction of the pessimism in the computed bounds on the other hand (see, for example, [32]).…”