DAG Tasks Research Articles

Time-Triggered Scheduling for Nonpreemptive Real-Time DAG Tasks Using 1-Opt Local Search

Time-Triggered Scheduling for Nonpreemptive Real-Time DAG Tasks Using 1-Opt Local Search

On the degree of parallelism for parallel real-time tasks

The degree of parallelism, which measures how a task can execute concurrently, is an important characterization in scheduling. This paper studies the degree of parallelism in the domain of real-time scheduling of parallel tasks, including the DAG task model and the conditional DAG task model. The definition of the degree of parallelism for DAG tasks is clarified; the definition and computing algorithm of the degree of parallelism for conditional DAG tasks are proposed. By leveraging the degree of parallelism, new response time bounds are derived and simple but effective real-time scheduling approaches are presented. This research is the first work to study the degree of parallelism for conditional DAG tasks and explore its benefits in real-time scheduling. Experimental results demonstrate that the proposed scheduling approaches significantly outperform existing state-of-the-art methods.

GA-DRL: Graph Neural Network-Augmented Deep Reinforcement Learning for DAG Task Scheduling Over Dynamic Vehicular Clouds

GA-DRL: Graph Neural Network-Augmented Deep Reinforcement Learning for DAG Task Scheduling Over Dynamic Vehicular Clouds

Priority-based DAG task offloading and secondary resource allocation in IoT edge computing environments

Priority-based DAG task offloading and secondary resource allocation in IoT edge computing environments

Efficient Workflow Scheduling in Edge Cloud-Enabled Space-Air-Ground-Integrated Information Systems

To address the challenges posed by the dynamism, high latency, and resource scarcity in integrated air-space-ground hybrid edge cloud environments on task completion times and node load, we designed a task scheduling system for scenarios involving the transmission and processing of interdependent tasks. This system integrates a graph neural network with attention mechanism and deep reinforcement learning. Specifically, we employ a graph encoder to extract features from DAG tasks and resources. Task scheduling solutions for dynamic environments are then generated using attention mechanism-equipped graph decoder, which are subsequently optimized based on performance metrics through the use of an Advantage Actor-Critic algorithm. Experimental results indicate that this algorithm performs well in terms of completion time and node load balance across tasks with different workflow structures, demonstrating its adaptability to highly dynamic edge cloud environments.

Research on Multi-DAG Satellite Network Task Scheduling Algorithm Based on Cache-Composite Priority

The problem of multiple DAGs sharing satellite constellation resources has gradually attracted widespread attention. Due to the limited computing resources and energy consumption of satellite networks, it is necessary to formulate a reasonable multi-DAG task scheduling scheme to ensure the fairness of each workflow under the premise of considering latency and energy consumption. Therefore, in this paper, we propose a multi-DAG satellite network task scheduling algorithm based on cache-composite priority under the Software-Defined Networking satellite network architecture. The basic idea of this algorithm lies in the DAG selection phase, where not only are task priorities computed but also the concept of fair scheduling is introduced, so as to prevent the excessively delayed scheduling of low-priority DAG tasks. In addition, the concept of public subtasks is introduced to reduce the system overhead caused by repetitive tasks. The experimental results show that the hybrid scheduling strategy proposed in this paper can meet the demand of DAG scheduling and improve the degree of task completion while effectively reducing the task latency and energy consumption.

Telemetry-aided cooperative multi-agent online reinforcement learning for DAG task scheduling in computing power networks

As demand for computing power and low latency in intelligence applications grows, the efficient management and coordination of resources in computing power networks become crucial. This paper presents a telemetry-aided multi-agent cooperation framework for DAG task scheduling in computing power networks. Utilizing distributed agents with network telemetry, the framework accurately assesses local network state information, formulates scheduling policies, and assigns tasks to edge servers. An online learning algorithm for DAG task scheduling is also introduced to enhance the cooperation strategy in decision-making, enabling rapid task scheduling and resource allocation decisions. Simulation results demonstrate a minimum 13.5% reduction in total task execution time compared to sub-optimal methods, along with improved node and link load balancing.

Online Offloading and Mobility Awareness of DAG Tasks for Vehicle Edge Computing

Online Offloading and Mobility Awareness of DAG Tasks for Vehicle Edge Computing

Design and Blocking Analysis of Locking Protocols for Real-Time DAG Tasks Under Federated Scheduling

Real-time systems require locking protocols to coordinate access to shared resources. With the booming revolution of parallel processing technology in real-time systems, there has been some work addressing the problem of extending classic locking protocols for sequential real-time tasks to parallel tasks. However, it may not be most effective to trivially follow the progress mechanisms and queue orders designed for sequential tasks since the intra-structure information within a parallel task is not taken into consideration. This paper investigates the design of locking protocols for parallel tasks using a novel mechanism -Longest Normal Section First (LNSF) -to consider the impact of normal sections on blocking behavior in parallel tasks and further improve real-time performance. LNSF is then implemented in a locking protocol for parallel tasks named POMIP, and associated blocking analysis techniques are presented. Empirical evaluations show that our proposed analysis dominated other state-of-the-art analysis-in best cases, the acceptance ratio of the task set can be improved by around 17%.

Real-Time Scheduling of Conditional DAG Tasks With Intra-Task Priority Assignment

The conditional DAG task model can represent the conditional execution flows that commonly exist in many real-time parallel applications. Previous work has shown that by properly assigning the priority among vertices inside a non-conditional DAG task, we can reduce the task response time and achieve better system schedulability. This paper studies how to apply intra-task priority assignment to conditional DAG tasks. We develop a response time bound that theoretically dominates the state-of-the-art and present a novel algorithm to compute the bound in polynomial time. We further extend the proposed approach to the general setting of multiple conditional DAG tasks. Experiments with one conditional DAG task and multiple conditional DAG tasks demonstrate that our method consistently outperforms the state-of-the-art by a considerable margin.

The Computational Complexity of Feasibility Analysis for Conditional DAG Tasks

The Conditional DAG (CDAG) task model is used for modeling multiprocessor real-time systems containing conditional expressions for which outcomes are not known prior to their evaluation. Feasibility analysis for CDAG tasks upon multiprocessor platforms is shown to be complete for the complexity class pspace ; assuming np ≠ pspace , this result rules out the use of Integer Linear Programming solvers for solving this problem efficiently. It is further shown that there can be no pseudo-polynomial time algorithm that solves this problem unless p = pspace .

RFID: Towards Low Latency and Reliable DAG Task Scheduling Over Dynamic Vehicular Clouds

Vehicular cloud (VC) platforms integrate heterogeneous and distributed resources of moving vehicles to offer timely and cost-effective computing services. However, the dynamic nature of VCs (i.e., limited contact duration among vehicles), caused by vehicles' mobility, poses unique challenges to the execution of computation-intensive applications/tasks with a directed acyclic graph (DAG) structure, where each task consists of multiple interdependent components (subtasks). In this paper, we study the scheduling of DAG tasks over dynamic VCs, where multiple subtasks of a DAG task are dispersed across vehicles and processed by vehicles cooperatively. We formulate DAG task scheduling as a 0-1 integer programming problem, aiming to minimize the overall task completion time while ensuring a high execution success rate, which turns out to be NP-hard. To tackle the problem, we develop a <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">r</u> anking and <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">f</u> oresight- <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i</u> ntegrated <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">d</u> ynamic scheduling scheme (RFID). RFID consists of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i)</i> a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">dynamic downward ranking</i> mechanism that sorts the scheduling priority of different subtasks, while explicitly taking into account the sequential execution nature of DAG; <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ii)</i> a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">resource scarcity-based priority changing</i> mechanism that overcomes possible performance degradations caused by the volatility of VC resources; and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">iii)</i> a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">degree-based weighted earliest finish time</i> mechanism that assigns the subtask with the highest scheduling priority to the vehicle which offers rapid task execution along with reliable transmission links. Simulation results reveal the effectiveness of our proposed scheme in comparison to benchmark methods.

Partial-order reduction in reachability-based response-time analyses of limited-preemptive DAG tasks

Response-time analysis (RTA) has been a means to evaluate the temporal correctness of real-time systems since the 1970 s. While early analyses were successful in capturing the exact upper bound on the worst-case response-time (WCRT) of systems with relatively simple computing platforms and task activation models, nowadays we see that most existing RTAs either become pessimistic or do not scale well as systems become more complex (e.g., parallel tasks running on a multicore platform). To make a trade-off between accuracy and scalability, recently, a new reachability-based RTA, called schedule-abstraction graph (SAG), has been proposed. The analysis is at least three orders of magnitude faster than other exact RTAs based on UPPAAL. However, it still has a fundamental limitation in scalability as it suffers from state-space explosion when there are large uncertainties in the timing parameters of the input jobs (e.g., large release jitters or execution-time variations). This could impede its applicability to large industrial use cases, or to be integrated with automated tools that explore alternative design choices. In this paper, we improve the scalability of the SAG analysis by introducing partial-order reduction rules that avoid combinatorial exploration of all possible scheduling decisions. We include systems with dependent and independent task execution models (i.e., with and without precedence constraint). Our empirical evaluations show that the proposed solution is able to reduce the runtime by five orders of magnitude and the number of explored states by 98% in comparison to the original SAG analysis. These achievements come only at a negligible cost of an over-estimation of 0.1% on the actual WCRT. We applied our solution on an automotive case study showing that it is able to scale to realistic systems made of hundreds of tasks for which the original analysis fails to finish.

Type-Aware Federated Scheduling for Typed DAG Tasks on Heterogeneous Multicore Platforms

To utilize the performance benefits of heterogeneous multicore platforms in real-time systems, we need task models that expose the parallelism and heterogeneity of the workload, such as typed DAG tasks, as well as scheduling algorithms that effectively exploit this information. In this paper, we introduce <i>type-aware federated scheduling</i> algorithms for sporadic typed DAG tasks with implicit deadlines running on a heterogeneous multicore platform with two different types of cores. In type-aware federated scheduling, a task can be executed in one of the three strategies: <i>Exclusive Allocation</i>, <i>Semi-Exclusive Allocation</i>, and <i>Sequential and Share</i>. In <i>Exclusive Allocation</i>, clusters of cores of both core types are exclusively allocated to tasks, while cores of only one type are exclusively allocated to tasks in <i>Semi-Exclusive Allocation</i>. The workload of the other type from tasks in <i>Semi-Exclusive Allocation</i> and the workload from tasks in <i>Sequential and Share</i> share the cores that are not exclusively allocated to any task. We prove that our type-aware federated scheduling algorithm has a capacity augmentation bound of 7.25. We also show that no constant capacity augmentation bound can be obtained without <i>Semi-Exclusive Allocation</i>. Compared to the state of the art, the type-aware federated scheduling algorithm achieves better schedulability, especially for task sets with skewed workload.

Segment based power-efficient scheduling for real-time DAG tasks on edge devices

Smart Mobile Devices (SMDs) are crucial for the edge computing paradigm’s real-world sensing. Real-time applications, which are computationally intensive and periodic with strict time constraints, can typically be used to replicate real-world sensing. Such applications call for increased processing speed, memory capacity, and battery life on SMDs, which are typically resource-constrained due to physical size restrictions. As a result, scheduling real-time applications for SMDs that are power efficient is crucial for the regular operation of edge computing platforms, and downstream decision-making tasks like computation offloading require the prediction of power consumption using power-saving approaches like DVFS. The main question is how to swiftly develop a better solution to the NP-Hard power efficient scheduling problem with DVFS. Thus, by segmenting the aligned tasks on an SMD, we present a segment-based analysis approach. Additionally, we offer a segment-based scheduling algorithm (SEDF) that draws inspiration from the segment-based analysis approach to achieve power-efficient scheduling for these real-time workloads. This segment-based approach yields a power consumption bound (PB), and a computation offloading use case is developed to demonstrate the application of PB in the subsequent decision-making processes. Both simulations and actual device tests are used to confirm the PB, SEDF, and the effectiveness of offloading decision-making. We demonstrate empirically that PB can be utilized to make approximative optimal decisions in decision-making problems involving computation offloading. SEDF is a straightforward and effective scheduling approach that can cut the power consumption of a multi-core SMD by roughly 30%.

The BH-mixed scheduling algorithm for DAG tasks with constrained deadlines

The development of real-time systems has stimulated studies of real-time scheduling algorithms and schedulability analysis methods for parallel tasks, whose number has increased from year to year. For parallel tasks that cannot be converted to sequential tasks, the existing scheduling strategies mostly use one scheduling strategy. From the experiment results when we use different scheduling strategies to schedule the same tasks, we can observe these strategies have advantages for tasks with specific parameter characteristics. We combine the strengths of three algorithms: the partitioned algorithm, the federated scheduling algorithm and the GFP algorithm. A BH-Mixed scheduling algorithm for directed acyclic graph tasks with constrained deadlines is proposed in this paper. We design a classification method according to the feature of these three algorithms. The tasks are divided into three groups and scheduled by the partitioned algorithm, the federated scheduling algorithm and the GFP algorithm respectively. Experimental results show that the BH-Mixed scheduling algorithm outperforms other algorithms tested in this paper.

The Bh-Mixed Scheduling Algorithm for Dag Tasks with Constrained Deadlines

The Bh-Mixed Scheduling Algorithm for Dag Tasks with Constrained Deadlines

Reliability-Aware Cost-Efficient Scientific Workflows Scheduling Strategy on Multi-Cloud Systems

Nowadays, more and more computation-intensive scientific applications with diverse needs are migrating to cloud computing systems. However, the cloud systems alone cannot meet applications’ requirements at all times with the increasing demands from users. Therefore, the multi-cloud systems that can provide scalable storage and computing resources become a good solution. The main challenges for such systems are multiple billing mechanisms, virtual resources heterogeneity, and systems reliability. In response to these challenges, we first build a multi-cloud systems fault-tolerant workflow scheduling framework, which tries to improve the scientific applications execution reliability and reduce their execution cost. Then, we use Weibull distribution to analyze task execution reliability and hazard rate, which is used to duplicate task with high execution hazard rate. Third, we integrate different multi-cloud providers’ billing mechanism into the proposed scheduling framework, and this workflow scheduling problem is mathematically formulated as an optimization problem. Fourth, we define the DAG tasks cost-efficient bottom level, and propose a fault-tolerant cost-efficient workflow scheduling algorithm (FCWS) that minimizes application execution cost, time while ensuring their reliability. Simulation experiments for performance evaluation were conducted based on two real-world applications: Epigenomics and LIGO. The results clearly demonstrate that our proposed FCWS algorithm outperforms existing FR-MOS, CWS in terms of cost and reliability, and FCWS is also better than CWS and inferior to FR-MOS in term of makespan.

Partitioning-Based Scheduling of OpenMP Task Systems With Tied Tasks

OpenMP is a popular programming framework in both general and high-performance computing and has recently drawn much interest in embedded and real-time computing. Although the execution semantics of OpenMP are similar to the DAG task model, the constraints posed by the OpenMP specification make them significantly more challenging to analyze. A tied task is an important feature in OpenMP that must execute on the same thread throughout its entire life cycle. A previous work [1] succeeded in analyzing the real-time scheduling of tied tasks by modifying the Task Scheduling Constraints (TSCs) in OpenMP specification. In this article, we also study the real-time scheduling of OpenMP task systems with tied tasks but without changing the original TSCs. In particular, we propose a partitioning-based algorithm, P-EDF-omp, by which the tied constraint can be automatically guaranteed as long as an OpenMP task system can be successfully partitioned to a multiprocessor platform. Furthermore, we conduct comprehensive experiments with both synthetic workloads and established OpenMP benchmarks to show that our approach consistently outperforms the work in [1] —even without modifying the TSCs.

A multiprocessor real-time scheduling embedded testbed based on Linux

We develop a testbed for multiprocessor real-time scheduling algorithms based on Linux and implement it on an Apalis iMX6 quad computer on module. This provides an interface to support different real-time task scheduling policies and a mechanism for managing the schedulers that can be used at runtime. Three built-in schedulers are currently available: GEDF, G-RM and DP-WRAP. With the proposed framework, periodic synchronous tasks with implicit deadlines can easily be generated using configuration files in JSON format, and parallel DAG tasks can be supported for G-EDF and G-RM, which can be used for evaluation of the schedulers. The open source tools trace-cmd and kernelshark are integrated into the testbed for task tracing and overhead recording. Unlike existing testbeds, ours is developed and tested based on the embedded platform, meaning that it is closer to a practical environment. Support for parallel DAG tasks is also a major feature of this testbed.