Composition of Schedulability Analyses for Real-Time Multiprocessor Systems

Lee, Jin-Kyu; Shin, Kang G.; Shin, Insik; Easwaran, Arvind

doi:10.1109/tc.2014.2308183

Cited by 18 publications

(9 citation statements)

References 35 publications

(92 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Indeed, when there is a significantly large parallel execution between the interfering task and the one under analysis, some sufficient tests can be so pessimistic that it is better to assume that the interfering task always runs upon a dedicated core. In Lee et al (2015) the performance of this compositional theory was studied with several specific schedulability tests. A similar method has also been used in Pathan and Jonsson (2014) to improve the priority assignment scheme in Davis and Burns (2011a) for G-FP scheduling.…”

Section: Related Workmentioning

confidence: 99%

Assessing the pessimism of current multicore global fixed-priority schedulability analysis

Sun

Natale

2018

Proceedings of the 33rd Annual ACM Symposium on Applied Computing

View full text Add to dashboard Cite

This work provides a formal assessment on the pessimism of existing methods for the schedulability analysis of multicores with global fixed priority (FP) scheduling. We show how according to the existing analysis methods for FP scheduling, it is relatively easy to define a simple task allocation strategy followed by local analysis that dominates the existing global-FP feasibility analysis algorithms, in terms of deadline guarantees. Rather than being an indication of a true comparison between the effectiveness of local and global policies, we consider the result as an indication of the limited maturity of multicore global analysis (and its outstanding pessimism). Additionally, we show how a simple change in the task model, consisting in splitting the task execution and performing the analysis in two stages, allows to provide a better global schedulability analysis that overcomes this limitation. However, the new analysis mainly aims to show where the demonstrated pessimism most likely comes from. Due to the huge pessimism pointed out on existing analyses and the limited improvement from the new analysis, more efforts from the community are needed to enhance the multicore global scheduling works.

show abstract

Section: Related Workmentioning

confidence: 99%

Assessing the pessimism of current multicore global fixed-priority schedulability analysis

Sun

Natale

2018

Proceedings of the 33rd Annual ACM Symposium on Applied Computing

View full text Add to dashboard Cite

show abstract

“…4 can be performed via three independent and identical analysis processes A, B and C. Each of the analysis processes is parametrized with WCIT. Moreover, the analysis processes can be performed in any order, though potentially in parallel Our framework provides the ability to analyze multicore systems, running any composition of scheduling algorithms, deemed to be non analyzable using regular schedulability tests [26]- [28].…”

Section: Conceptual Design 31mentioning

confidence: 99%

Compositional Schedulability Analysis of Multicore Modular Avionic Architectures

Boudjadar¹

2018

JCP

View full text Add to dashboard Cite

This paper presents a compositional schedulability analysis of multicore modular avionic systems (IMA). It provides a fine grained description of the software architecture and behavior to reduce the over-approximation, and a holistic analysis to check schedulability while considering computation requirements and shared memory interference. The system is structured in terms of subsystems, encapsulating ARINC653 partitions, each of which runs on a processing core. The schedulability analysis is performed for each subsystem individually while accounting for the memory interference that would result if the core under analysis runs effectively alongside with the rest of cores. Thereafter, we introduce an architecting technique to relocate functions between tasks located at the same partition in case of non schedulability, to derive potential schedulable system configurations delivering the same functionality. Schedulability is formally analyzed using Uppaal model checker. Our evaluation results show that our compositional analysis technique consumes up to 95% less than regular analysis, in terms of analysis time and memory space. 1202 perfect memory access i.e., the memory is immediately available whenever an access request occurs (hits) [6]- [9]. Such an estimation of the WCET at task level is most likely under-approximating. We believe that estimating the execution time at a lower level of the granularity (e.g. function level) results in less underapproximation of the execution time. Thus, the more granular the behavior representation is, the more optimistic the WCET will be. The under-approximation difference between task and function levels could be non comparable and would increase drastically with fine grained description levels.The interference resulting from shared memories and communication means (buses, networks) is a determinant factor in the schedulability of component-based real-time system. The interference time is in fact related to the number of concurrent components and the bandwidth of shared resources. Accordingly, accurate schedulability techniques require a holistic analysis where both computation and memory interference/ communication have to be considered together [10], [5].A surge of progress has been achieved in the area of schedulability analysis of multicore systems through an intensive use of model-based settings and formal methods. However, due to the systems size in avionics, given by the integration of huge number of concurrent applications, the use of formal methods can end up in state space explosion. Different techniques have been considered to bypass the state space explosion and provide upper bound guarantees on the schedulability, we cite abstraction-based [11], compositional analysis [12] and incremental analysis [13]. In this paper, we introduce a model-based framework for fine grained modeling and formal schedulability analysis of multicore avionic (IMA-driven) systems with shared memories. The system is structured in terms of subsystems, encapsulating ARINC653 partitions, each ...

show abstract

“…As a basis for task set generation, we adapt the technique in [33], which has been widely used [34,35]. To broaden the spectrum of task sets to be tested, we use 10 different distributions for C i /T i : bimodal distributions with p =0.1, 0.3, 0.5, 0.7 or 0.9, where the value of C i /T i is uniformly chosen in [0, 0.5) with probability p and in [0.5, 1) with probability 1 − p, and exponential distributions with 1/λ =0.1, 0.3, 0.5, 0.7 or 0.9, whose probability density function is λ • exp(−λ • x).…”

Section: Task Set Generationmentioning

confidence: 99%

Development and use of a new task model for cyber-physical systems: A real-time scheduling perspective

Lee

Shin

2017

Journal of Systems and Software

Self Cite

View full text Add to dashboard Cite

In a typical cyber-physical system (CPS), the cyber/computation subsystem controls the physical subsystem, and therefore the computer society has recently paid considerable attention to CPS research. To keep such a CPS stable, feedback control with periodic computation tasks has been widely used, and its theoretical guarantee of stability has been made with periodic real-time task models that enforce strict periodic control updates. However, some control update misses are usually allowed (e.g., via system over-design) in certain physical subsystem states (PSSes) without causing system instability, and the resources required for strict periodic control updates can thus be reduced or used for other purposes, achieving efficient controls for the entire CPS in terms of the operational cost, such as fuel consumption or tracking accuracy. In this paper, we propose a new periodic, fault-tolerant CPS task model, which not only expresses efficiency and stability of the underlying physical subsystem, but also generalizes existing periodic real-time task models, by capturing a tolerable number of control update misses in different PSSes. To demonstrate the utility of this model, we develop a new scheduling mechanism that prioritizes jobs (i.e., periodic invocations) of a set of tasks not only by the nature of each task, but also by the number of consecutive prior job deadline misses. Based on its analysis in terms of stability and efficiency, we also propose a priority-assignment policy that lowers the system operation cost without compromising stability. Our in-depth analysis and simulation results show that the scheduling mechanism and its analysis, as well as the priorityassignment policy under the proposed model not only generalize the existing periodic real-time task models, but also significantly lower the system operation cost without losing stability.

show abstract

Composition of Schedulability Analyses for Real-Time Multiprocessor Systems

Cited by 18 publications

References 35 publications

Assessing the pessimism of current multicore global fixed-priority schedulability analysis

Assessing the pessimism of current multicore global fixed-priority schedulability analysis

Compositional Schedulability Analysis of Multicore Modular Avionic Architectures

Development and use of a new task model for cyber-physical systems: A real-time scheduling perspective

Contact Info

Product

Resources

About