Best-Case Response Times and Jitter Analysis of Real-Time Tasks.
J. Scheduling 01/2004; 7:133-147. pp.133-147
Article: Exact best-case response time analysis of real-time tasks under fixed-priority pre-emptive scheduling for arbitrary deadlines[show abstract] [hide abstract]
ABSTRACT: In this paper, we present a conjecture for exact best-case response times of periodic released, independent real-time tasks with arbitrary deadlines that are scheduled by means of fixed-priority pre-emptive scheduling (FPPS). We illus-trate the analysis by means of an example. Apart from hav-ing a value on its own whenever timing constraints include lower bounds on response times of a system to events, the novel analysis allows for an improvement of existing end-to-end response time analysis in distributed systems, i.e. where the finalization of one task on a processor activates another task on another processor.
Article: Towards pragmatic solutions for two-level hierarchical scheduling Part I: A basic approach for independent applications[show abstract] [hide abstract]
ABSTRACT: Resource reservation has been proposed in the literature as a means to prevent temporal interference between applica-tions. With applications consisting of one or more real-time tasks, resource reservation inherently involves multiple levels of scheduling, i.e. a scheduling hierarchy. This document considers two-level hierarchical scheduling for independent applications in cost-constrained systems, using fixed-priority pre-emptive scheduling (FPPS) for tasks. We aim at pragmatic solutions for such systems, which allow efficient implementations of budgets and simple analysis for tasks. Given the complexity and pessimism of the analysis for tasks for existing approaches and based on the observation that budgets are design artifacts, we consider a basic approach, where budgets are time-triggered and all budgets have the same period. Using jitter analysis as an example, it is shown that existing analytic results for real-time tasks under FPPS can be easily converted to two-level hierarchical scheduling for this approach by viewing the unavailability of a budget as an artificial highest priority task. We show that the conversion equally well applies to the approaches described in [12, 17, 18], amongst others. We briefly consider an advanced approach where budgets have harmonic periods, and show that the conversion is not straightforward for such an approach.
[show abstract] [hide abstract]
ABSTRACT: Given a set of applications, each application is assigned a quality level so that the overall output quality is maximized. For each quality level, the applications are assigned resource budgets (similar to reservations in ). Initially, these budgets are based on resource estimates, but that can be adapted dynamically to adjust to the real needs of the application. The resources budgets contribute to robustness by preventing temporal interference between applications. However, for cost-effectiveness reasons, the budget is chosen below worst-case, giving rise to overload situations. We describe a design for conditionally guaranteed budgets (CGBs) and initial accompanying timing analysis. Our work is carried out in the context of a resource management approach for media processing in consumer terminals, where applications can run at different quality levels and receive budgets that are both guaranteed and enforced. Distinguishing characteristics of our work are: (1) the aim for stable output quality of applications, since frequent changes of output quality are perceived as non-quality, (2) budgets below worst-case, for cost-effectiveness reasons, giving rise to overload situations, and (3) different solutions for transient and structural overloads. Figure 1 shows load variations and resource budgets of a media application. The budget covers most of the load variations around the average, and the we assume that the applications resolve the transient overloads, indicated by dots in the figure. This figure also depicts, at time t i , a structural load increase, i.e. a load variation that increases the average load. For stability reasons, the system has to distinguish structural from transient overloads, and react only to the former one. Therefore, the structural nature of the load increase has to be detected first. Next, the new set of quality levels and budgets is determined, and finally the adaptation takes place, at time t a. During this time (t a -t i) the media application has a budget shortage, and its output quality is degraded. In this paper, we focus on the system aspects of the structural overloads. CGB were conceived as a solution for cost-effectiveness and stability in the context of structural overloads. The contribution of this paper is a design of CGBs where the CGB is consumed by the receiving application in the place of another application that consistently uses less than its guaranteed budget. Furthermore, we present initial analytical results based on best-case and worst-case analysis.01/2001;
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.