Prediction of end-to-end deadline missing in distributed threads systems

Abstract

Distributed real-time threads are schedulable entities with an end-to-end deadline that traverse nodes, carrying their scheduling context. In each node, the thread will be locally scheduled and predictions about deadline missing allow that actions are carried out to improve system performance. This paper presents a task model and deadline partitioning algorithms that consider the possibility of a distributed thread to follow different paths. The future execution flow of the distributed thread is only probabilistic known before its execution. End-to-end deadline missing prediction mechanisms can be carried out through definition of estimated local deadlines. Simulations show that the proposed prediction mechanism presents good results in overloaded systems.

Deadline Missing Prediction in Systems Based on Distributed Threads

“…In [13] it is proposed a deadline missing prediction mechanism which uses deadline partitioning methods (ED, EQS and EQF) in order to define target local response times, called milestones. These milestones are used as response time predictors.…”

“…Simulations were carried out with the objective of comparing the performance of the ARS mechanism with Milestone mechanism proposed in [13]. The Relative Error Rate (E(z)) and Correct Prediction Rate (CP(z)) metrics were used to compare the results of the proposed mechanisms.…”

“…Deadline partitioning methods are applied at the creation of each DT activation, before it begins to execute (static way). Method EQF [13] is used in this work to partition deadlines according to some DT itinerary.…”

“…These itineraries are kept stored in the DT's log. Although a DT is of an autonomous nature and it is difficult to predict the next node that it will traverse, through the observation of its historical data it is possible to define four main itineraries that a DT would follow [13]:…”