An ILP-based Worst-case Performance Analysis Technique for Distributed Real-time Embedded Systems

Abstract

Abstract-Finding a tight upper bound of the worst-case response time in a distributed real-time embedded system is a very challenging problem since we have to consider execution time variations of tasks, jitter of input arrivals, scheduling anomaly behavior in a multi-tasking system, all together. In this paper, we translate the problem as an optimization problem and propose a novel solution based on ILP (Integer Linear Programming). In the proposed technique, we formulate a set of ILP formulas in a compositional way for modeling flexibility, but solve the problem holistically to achieve tighter upper bounds. To mitigate the time complexity of the ILP method, we perform static analysis based on a scheduling heuristic to reduce the number of variables and confine the variable ranges. Preliminary experiments with the benchmarks used in the related work and a real-life example show promising results that give tight bounds in an affordable solution time.

SimLatte: A Framework to Support Testing for Worst-Case Interrupt Latencies in Embedded Software

“…There has been a great deal of work on verifying temporal properties of embedded software using model checking [9], [16], static analysis [21], [30], and testing [22], [24]. For example, Hessel et al [16] use UPPAAL, a model checker, to verify embedded systems that can be modeled as timed automata against their real time specifications.…”

“…While many advances have been made in the area of realtime system verification, particularly in the area of static analysis (e.g., [9], [16], [21], [22], [30]), the correctness of interrupt-driven systems remains difficult to verify. Without running the system being assessed, it is difficult for static analysis to identify locations where interrupts can occur because interrupts are platform dependent, and can be disabled and enabled by other software components or underlying systems at arbitrary execution points.…”

Improved analysis and evaluation of real-time semaphore protocols for P-FP scheduling

“…In very recent work, Kim et al [18] applied linear programming to the task of determining tight response-time bounds in distributed systems under P-FP scheduling with precedence constraints (and without locking). In contrast to our LP-based, approach, in which a (not necessarily tight) upper bound on pi-blocking is expressed by means of non-integral blocking fractions, Kim et al use integer linear programs (ILPs) to find schedules that maximize response times.…”

A novel analytical method for worst case response time estimation of distributed embedded systems

“…Recently, an analysis technique based on integer linear programming (ILP) was proposed to estimate the worst-case performance of a nonpreemptive multitasking MPSoC [2]. The technique proposed by Madl [9] enumerates all possible event execution paths and verifies them using discrete event simulation to analyze the schedulability of a distributed embedded system assuming nonpreemptive scheduling.…”

“…A model checking approach [1] and an ILP-based approach [2] have been proposed to accurately estimate the WCRT. But they both suffer from poor scalability since the time complexity is inherently exponential.…”