Loading...
3 results
Search Results
Now showing 1 - 3 of 3
- Response time analysis of multiframe mixed-criticality systems with arbitrary deadlinesPublication . Hussain, Ishfaq; Awan, Muhammad Ali; Souto, Pedro; Bletsas, Konstantinos; Akesson, Benny; Tovar, EduardoThe well-known model of Vestal aims to avoid excessive pessimism in the quantifcation of the processing requirements of mixed-criticality systems, while still guaranteeing the timeliness of higher-criticality functions. This can bring important savings in system costs, and indirectly help meet size, weight and power constraints. This efciency is promoted via the use of multiple worst-case execution time (WCET) estimates for the same task, with each such estimate characterized by a confdence associated with a diferent criticality level. However, even this approach can be very pessimistic when the WCET of successive instances of the same task can vary greatly according to a known pattern, as in MP3 and MPEG codecs or the processing of ADVB video streams. In this paper, we present a schedulability analysis for the new multiframe mixed-criticality model, which allows tasks to have multiple, periodically repeating, WCETs in the same mode of operation. Our work extends both the analysis techniques for Static Mixed-Criticality scheduling (SMC) and Adaptive Mixed-Criticality scheduling (AMC), on one hand, and the schedulability analysis for multiframe task systems on the other. A constrained-deadline model is initially targeted, and then extended to the more general, but also more complex, arbitrary-deadline scenario. The corresponding optimal priority assignment for our schedulability analysis is also identifed. Our proposed worst-case response time (WCRT) analysis for multiframe mixed-criticality systems is considerably less pessimistic than applying the static and adaptive mixed-criticality scheduling tests oblivious to the WCET variation patterns. Experimental evaluation with synthetic task sets demonstrates up to 20% and 31.4% higher scheduling success ratio (in absolute terms) for constrained-deadline analyses and arbitrary-deadline analyses, respectively, when compared to the best of their corresponding frame-oblivious tests.
- Schedulability analysis for CAN bus messages of periodically-varying sizePublication . Hussain, Ishfaq; Souto, Pedro; Bletsas, Konstantinos; Awan, Muhammad Ali; Tovar, EduardoConventional CAN bus schedulability analysis assumes that all messages with a given identifier have the same worst-case length. In this paper we extend that analysis to a more general model in which messages with a given identifier may have different lengths, that vary according to a known periodic pattern.That is, for some positive integer S, we assume that the length of message instances n and n + S with the same id is the same. By leveraging such patterns, where present, our new analysis allows for a more efficient use of CAN bus bandwidth than the application of conventional analysis, which can be pessimistic. This may be interesting when a given node sends the values of multiple signals with different periods. In such a scenario, the conventional CAN schedulability analysis would require either the use of different ids for different signals (assuming there are enough of them), which leads to a higher bandwidth overhead because of the reduplication of message headers, or using only one id, but pessimistically always assuming the maximum possible length of the message, for safety reasons.
- Response time analysis of memory-bandwidth- regulated multiframe mixed-criticality systemsPublication . Hussain, Ishfaq; Awan, Muhammad Ali; Souto, Pedro; Bletsas, Konstantinos; Tovar, EduardoThe multiframe mixed-criticality task model eliminates the pessimism in many systems where the worst-case execution times (WCETs) of successive jobs vary greatly by design, in a known pattern. Existing feasibility analysis techniques for multiframe mixed-criticality tasks are shared-resource-oblivious, hence un-safe for commercial-o -the-shelf (COTS) multicore platforms with a memory controller shared among all cores. Conversely, the feasibility analyses that account for the interference on shared resource(s) in COTS platforms do not leverage theWCET variation in multiframe tasks. This paper extends the state-of-the-art by presenting analysis that incorporates the memory access stall in memory-bandwidth-regulated multiframe mixed-criticality multicore systems. An exhaustive enumeration approach is proposed for this analysis to further enhance the schedulability success ratio. The running time of the exhaustive analysis is improved by proposing a pruning mechanism that eliminates the combinations of interfering job sequences that subsume others. Experimental evaluation, using synthetic task sets, demonstrates up to 72% improvement in terms of schedulability success ratio, compared to frame-agnostic analysis.