In this chapter, we describe a stochastic analysis framework that does not introduce any worst-case orrestrictive assumptions into the analysis, and is applicable to general priority-driven real-time systems. Theproposed framework builds up on Stochastic Time Demand Analysis (STDA) in that the techniques used inthe framework to compute the response time distributions of tasks are largely borrowed from the STDA.However, unlike the STDA, which focuses on particular execution scenarios starting at a critical instant,the proposed framework considers all possible execution scenarios in order to obtain the exact response time distributions of the tasks. Moreover, while the STDA addresses only fixed-priority systems such as RateMonotonic and Deadline Monotonic, our framework extends to dynamic -priority systems such asEarliest Deadline First.
Stochastic Analysis of PriorityDriven Periodic RealTime Systems
LO BELLO, Lucia;MIRABELLA, Orazio
2007-01-01
Abstract
In this chapter, we describe a stochastic analysis framework that does not introduce any worst-case orrestrictive assumptions into the analysis, and is applicable to general priority-driven real-time systems. Theproposed framework builds up on Stochastic Time Demand Analysis (STDA) in that the techniques used inthe framework to compute the response time distributions of tasks are largely borrowed from the STDA.However, unlike the STDA, which focuses on particular execution scenarios starting at a critical instant,the proposed framework considers all possible execution scenarios in order to obtain the exact response time distributions of the tasks. Moreover, while the STDA addresses only fixed-priority systems such as RateMonotonic and Deadline Monotonic, our framework extends to dynamic -priority systems such asEarliest Deadline First.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.