Managing latency in embedded streaming applications under hard-real-time scheduling

In this paper, we consider the problem of hard-real-time scheduling of embedded streaming applications, modeled using dataflow graphs, while minimizing the application latency. Recently, it has been shown that the actors in an acyclic Cyclo-Static Dataflow (CSDF) graph can be scheduled as a set of implicit-deadline periodic tasks. Such scheduling approach has been shown to yield the maximum achievable throughput for a large set of graphs, called matched I/O rates graphs. We show that scheduling the graph actors as implicit-deadline periodic tasks increases the latency significantly for a class of graphs called unbalanced graphs. To alleviate this problem, we propose a new task-set representation for the actors in which the actors are scheduled as a set of constrained-deadline periodic tasks. We prove that scheduling the actors as constrained-deadline periodic tasks delivers optimal throughput (i.e., rate) and latency for graphs with repetition vector equal to $\vec{1}$. Furthermore, we evaluate the constrained-deadline representation using a set of 19 real-life applications and show that it is capable of achieving the minimum achievable latency for more than 70% of the applications, and even if the application has a repetition vector not equal to $\vec{1}$. We show that choosing the task deadline involves a trade-off between the latency and the resources requirements. Finally, we propose a decision tree to assist the designer in choosing the appropriate real-time periodic task model for scheduling acyclic CSDF graphs.