Improved Schedulability Analysis of EDF on Multiprocessor Platforms

A model is considered for representing recurrent precedence-constrained tasks that are to execute on multiprocessor platforms. A recurrent task is specified as a directed acyclic graph (DAG), a period, and a relative deadline. Each vertex of the DAG represents a sequential job, while the edges of the DAG represent precedence constraints between these jobs. All the jobs of the DAG are released simultaneously and need to complete execution within the specified relative deadline of their release. Each task may release jobs in this manner an unbounded number of times, with successive releases occurring at least the specified period apart. Conditional control structures are also allowed. The scheduling problem is to determine whether a set of such recurrent tasks can be scheduled to always meet all deadlines upon a specified number of identical processors. This problem is shown to be computationally intractable, but amenable to efficient approximate solutions. Earliest Deadline First (EDF) and Deadline Monotonic (DM) are shown to be good approximate global scheduling algorithms. Polynomial and pseudo-polynomial time schedulability tests, of differing effectiveness, are presented for determining whether a given task set can be scheduled by EDF or DM to always meet deadlines on a specified number of processors.

A new technique was recently introduced by Bonifaci et al. for the analysis of real-time systems scheduled on multiprocessor platforms by the global Earliest Deadline First (EDF) scheduling algorithm. In this paper, this technique is generalized so that it is applicable to the schedulability analysis of real-time systems scheduled on multiprocessor platforms by any work-conserving algorithm. The resulting analysis technique is applied to obtain a new sufficient global Deadline Monotonic (DM) schedulability test. It is shown that this new test is quantitatively superior to pre-existing DM schedulability analysis tests; in addition, the degree of its deviation from any hypothetical optimal scheduler (that may be clairvoyant) is quantitatively bounded. A new global EDF schedulability test is also proposed here that builds on the results of Bonifaci et al. This new test is shown to be less pessimistic and more widely applicable than the earlier result was, while retaining the strong theoretical properties of the earlier result.

We address the problem of schedulability analysis for a set of sporadic real-time tasks scheduled by the Global Earliest Deadline First (G-EDF) policy on a multiprocessor platform. State-of-the-art tests for schedulability analysis of multiprocessor global scheduling are often incomparable. That is, a task set that is judged not schedulable by a test may be verified to be schedulable by another test, and vice versa. In this paper, we first develop a new schedulability test that integrates the limited carry-in technique and Response Time Analysis (RTA) procedure for Global EDF schedulability analysis. Then, we provide an over-approximate variant of this test with better run-time efficiency. Later, we extend these two tests to self-suspending tasks. All schedulability tests proposed in the paper have provable dominance over their state-of-the-art counterparts. Finally, we conduct extensive comparisons among different schedulability tests. Our new tests show significant improvements for schedulability analysis of Global EDF.

Many algorithms have recently been studied for scheduling mixed-criticality (MC) tasks. However, most existing MC scheduling algorithms guarantee the timely executions of high-criticality (HC) tasks at the expense of discarding low-criticality (LC) tasks, which can cause serious service interruption for such tasks. In this work, aiming at providing guaranteed services for LC tasks, we study an elastic mixed-criticality (E-MC) task model for dual-criticality systems. Specifically, the model allows each LC task to specify its maximum period (i.e., minimum service level) and a set of early-release points. We propose an early-release (ER) mechanism that enables LC tasks to be released more frequently and thus improve their service levels at runtime, with both conservative and aggressive approaches to exploiting system slack being considered, which is applied to both earliest deadline first (EDF) and preference-oriented earliest-deadline schedulers. We formally prove the correctness of the proposed early-release--earliest deadline first scheduler on guaranteeing the timeliness of all tasks through judicious management of the early releases of LC tasks. The proposed model and schedulers are evaluated through extensive simulations. The results show that by moderately relaxing the service requirements of LC tasks in MC task sets (i.e., by having LC tasks’ maximum periods in the E-MC model be two to three times their desired MC periods), most transformed E-MC task sets can be successfully scheduled without sacrificing the timeliness of HC tasks. Moreover, with the proposed ER mechanism, the runtime performance of tasks (e.g., execution frequencies of LC tasks, response times, and jitters of HC tasks) can be significantly improved under the ER schedulers when compared to that of the state-of-the-art earliest deadline first—virtual deadline scheduler.

In this paper we show how to extend classical real time feasibility conditions to take into account kernel overheads for mixed preemptive and non- preemptive periodic tasks with Earliest Deadline First (EDF) scheduling policy. The EDF algorithm allows us to schedule some tasks sets which are not valid with FP/FIFO scheduling policy, the classical scheduling implemented for OSEK. That is why we have improved our OSEK by implementing EDF algorithm. First of all, we identify the sources of overheads that influence the response time of the tasks: the context switching, the mechanisms used to activates/terminates and reschedules tasks, and the granularity of the periodic timer used to implement the periodic task model. We then show how to take into account those overheads in the classical EDF feasibility conditions. We compare the response times computed with theoretical feasibility conditions with kernel overhead to the results obtained on a real implementation. We show that the kernel overheads cannot be neglected and that our theoretical results are valid and can be used for a real-time dimensioning of an OSEK system with EDF scheduling policy. Finally, we compare the performances of our EDF implementation to the FP/FIFO native scheduling showing the benefits of implementing EDF in OSEK.

Internet of Things (IoT) based healthcare applications are time‐sensitive and any delay can cause alarming situations, including death of patients. The Earliest Deadline First (EDF) scheduling scheme has been proposed for use in IoT‐based healthcare applications. However, the EDF scheme performs poorly under overloaded conditions due to giving highest priority to packets that are close to missing their deadlines. Some studies have proposed the use of Priority EDF to overcome the challenges of EDF; however, Priority EDF still favours higher priority queues which increases the waiting times of lower priority queues. In order to overcome the limitation of EDF and its variants, this paper proposes a system model for a prioritized scheduling (PS) scheme. The PS scheme is an improvement of the Earliest Deadline First (EDF) scheme and its variants for IoT‐based healthcare applications. The PS scheme uses a heterogeneous multi‐server priority queuing system to provide service differentiation by prioritizing short packets over large packets and delay sensitive packets are serviced before delay tolerant packets. Numerical results demonstrate that the PS scheme minimizes the mean slowdown for both delay sensitive short and large packets at low and high load values. Additionally, the PS scheme performs better than the EDF and Priority EDF schemes in terms of reducing mean slowdown of packets and the PS scheme performs better than the EDF in terms of throughput for all packet sizes at both low and high load values. The performance improvement in terms of throughput is more pronounced at high load values. This addresses the challenge of the EDF scheme which performs poorly under overloaded conditions and the challenge of the Priority EDF scheme which favours higher priority queues at the expense of low priority queues.

In G. C. Buttazzo, Real-Time Syst., vol. 29, no. 1, pp. 5-26, 2005, the author empirically compared earliest deadline first (EDF) and rate monotonic (RM) scheduling algorithms and made a few EDF preemption behavior observations based on data obtained from the first 1000 time units of scheduling activities. However, based on test settings given in (G. C. Buttazzo, Real-Time Syst., vol. 29, no. 1, pp. 5-26, 2005), the first 1000 time units occupies only a small percentage of the entire task set's hyper-period. We extend EDF preemption behavior study by extending scheduling activities from the first small percentage of a hyper-period of a given task to the entire hyper-period. The extended experiments indicate that the number of preemptions occurred at the beginning of a task set's hyper-period does not necessarily represent the trend for the entire hyper-period. Hence, comparisons and conclusions made based on a small percentage of a scheduling interval over a task set's hyper-period may not be accurate. Second, the total number of preemptions within a task set's hyper-period does not decrease when the task set total utilization increases which is different from the observation obtained in (G. C. Buttazzo, Real-Time Syst., vol. 29, no. 1, pp. 5-26, 2005). We also investigate the impact of execution time differences among tasks on the preemption behavior.

Cache partitioning is a technique to reduce interference among tasks running on the processors with shared caches. To make this technique effective, cache segments should be allocated to tasks that will benefit the most from having their data and instructions stored in the cache. The requests for cached data and instructions can be retrieved faster from the cache memory instead of fetching them from the main memory, thereby reducing overall execution time. The existing partitioning schemes for real-time systems divide the available cache among the tasks to guarantee their schedulability as the sole and primary optimization criterion. However, it is also preferable, particularly in systems with power constraints or mixed criticalities where low- and high-criticality workloads are executing alongside, to reduce the total cache usage for real-time tasks. Cache minimization as part of design space exploration can also help in achieving optimal system performance and resource utilization in embedded systems. In this paper, we develop optimization algorithms for cache partitioning that, besides ensuring schedulability, also minimize cache usage. We consider both preemptive and non-preemptive scheduling policies on single-processor systems with fixed- and dynamic-priority scheduling algorithms (Rate Monotonic (RM) and Earliest Deadline First (EDF), respectively). For preemptive scheduling, we formulate the problem as an integer quadratically constrained program and propose an efficient heuristic achieving near-optimal solutions. For non-preemptive scheduling, we combine linear and binary search techniques with different fixed-priority schedulability tests and Quick Processor-demand Analysis (QPA) for EDF. Our experiments based on synthetic task sets with parameters from real-world embedded applications show that the proposed heuristic: (i) achieves an average optimality gap of 0.79% within 0.1× run time of a mathematical programming solver and (ii) reduces average cache usage by 39.15% compared to existing cache partitioning approaches. Besides, we find that for large task sets with high utilization, non-preemptive scheduling can use less cache than preemptive to guarantee schedulability.

Deriving deadlines and periods for update transactions so as to maintain timeliness and data freshness while minimizing imposed workload has long been recognized an important problem in real-time database research. Despite years of active research, the state-of-the-art still has much room for improvement, particularly for periodic transactions scheduled by the Earliest Deadline First (EDF) algorithm. In this paper, we propose a practical and efficient two-phase algorithm, GEneral EDF (GE <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">EDF</sub> ), for assigning periods and deadlines to EDF-scheduled update transactions. Phase 1 of GE <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">EDF</sub> aims at finding solutions for most inputs in linear time, based on the observation that the execution times of update transactions are relatively small compared to the validity interval lengths of real-time data objects in many real-time applications. In the remaining cases for which Phase 1 fails to derive solutions, Phase 2 is invoked by employing an existing deadline-monotonic-based algorithm, which we show is also applicable to our problem. Meanwhile, we have devised several techniques which significantly reduce the cost of schedulability test, and hence greatly improve time efficiency. Our experimental results demonstrate that GE <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">EDF</sub> outperforms existing approaches in terms of generated workloads. Although Phase 2 has a pseudopolynomial time complexity, our experimental study shows that it runs much faster than other solutions with comparable quality.

The Earliest Deadline First scheduling algorithm (EDF) is known to not be optimal under global scheduling on multiprocessor platforms. Results have been obtained that bound the maximum tardiness&#x2013; the amount of time by which deadlines may be missed&#x2013; of any feasible system of implicit-deadline sporadic tasks scheduled using global EDF. However, it is known that these bounds are not tight. In this paper, we derive an algorithm for obtaining tardiness bounds that are superior to previously known bounds. In contrast to prior algorithms, which compute a single tardiness bound for all the tasks in the system, our algorithm derives a separate tardiness bound for each task. Particularly upon task systems in which the parameters of the tasks are very dissimilar, our bounds are significantly better than prior bounds. Our algorithm also yields a simple sufficient test for the tardiness verification problem: given a task system with maximum acceptable tardiness bounds per task, is the system guaranteed to be scheduled by global EDF such that these tardiness constraints are not violated?

On the computational complexity of periodic scheduling

Considering the execution order restriction of control tasks and message tasks in asynchronous networked control system (NCS), this paper proposes a hybrid scheduling scheme based on earliest deadline first (EDF) algorithm and genetic algorithm (GA). GA is used to allocate control tasks on multiprocessors in order to achieve load balance. For the purpose of meeting message deadlines and maximizing network utilization, non-preemptive EDF is used for the transmission of messages on networks. Finally, a tasks and messages scheduling paradigm is constructed within a least common multiple (LCM) which contains information about execution time and resource allocation of messages and tasks. Simulation results show that execution order restriction and timing constraints can be satisfied and a considerable high resource utilization rate can be achieved through our scheduling scheme.

EDF (Earliest Deadline First) has been proved to be optimal scheduling algorithm for single processor real-time system. It also performs well for multiprocessor system. Limitation of EDF is that its performance decreases exponentially when system becomes slightly overloaded. ACO (Ant Colony Optimization) based scheduling algorithm performs well in both underloaded and overloaded conditions. But its limitation is that it takes more time for execution compared to EDF. In this paper, an adaptive algorithm for multiprocessor real-time system is proposed, which is combination of both of these algorithms. The proposed algorithm along with EDF and ACO based algorithm is simulated for real-time multiprocessor system and the results are obtained. The performance is measured in terms of Success Ratio (SR) and Effective CPU Utilization (ECU). Execution Time taken by each scheduling algorithm is also measured. From analysis and experiments, it reveals that the proposed algorithm is fast as well as efficient in both underloaded and overloaded conditions for real-time multiprocessor systems.

In order to analyze and evaluate real-time performance of CAN network, the OPNET models of CAN network with Fixed Priority Scheduling (FPS) and Earliest Deadline First (EDF) have been set up in this paper. Based on simulation experiments, the performances of average delay and transmission success ratio in the two models are analyzed and the experiments also indicate that under normal load condition the two scheduling algorithm both can send data frame successfully before the deadline. In a slight overload condition FPS algorithm makes the transmission delay of low priority data frame significantly increased, and even occurs the phenomenon that low priority data frame can not be sent at all times; while the EDF algorithm can make all the nodes to complete the data frames in less than 3ms transmission delay, it meets the delay requirements that are delay should less than 10ms in most control systems, it indicates that EDF algorithm has the better scheduling and availability in soft real-time system.

We present a task model for the real-time execution of event-driven tasks in which no a priori characterization of the actual arrival rates of events is known; only the expected arrival rates of events is known. The model, called rate-bared execution (RBE), is a generalization of Mok's sporadic task model. The RBE model is motivated naturally by distributed multimedia and digital signal processing applications. We derive necessary and sufficient conditions for determining the feasibility of an RBE task set and demonstrate that earliest deadline first (EDF) scheduling is an optimal scheduling algorithm for both preemptive and nonpreemptive execution environments, as well as hybrid environments wherein RBE tasks access shared resources. Our analysis of RBE tasks demonstrates a fundamental distinction between deadline based scheduling methods and static priority based methods. We show that for deadline-based scheduling methods, feasibility is solely a function of the distribution of task deadlines in time. This is contrasted with static priority schedulers where feasibility is a function of the actual arrival rates of work for tasks. Thus whereas the feasibility of static priority schedulers is a function of the periodicity of tasks, the feasibility of deadline schedulers is independent of task arrival processes and hence deadline schedulers are more suitable for use in distributed, event-driven, real-time systems.