Directed Acyclic Graph Tasks Research Articles

A method for simultaneously implementing trajectory planning and DAG task scheduling in multi-UAV assisted edge computing

UAV-assisted edge computing(UEC) as a new framework is able to provide computing services to remote areas. However, facing computationally intensive tasks with huge computation time forces them to hover near the user’s devices(UDs) for long periods of time. To better utilize the available arithmetic resources and reduce the computation time of UAVs, it is imperative to introduce directed acyclic graph (DAG) task scheduling into the UEC framework. Therefore, this article proposes a DAG-type task-driven trajectory planning (DAG-TDTP) model, which can plan UAV routes while scheduling DAG subtasks between UAVs that offload from UDs. To implement the DAG-TDTP model, we propose a distance-based heterogeneous earliest-finish-time (D-HEFT) algorithm and a time segmentation method based on the cooperative task offloading matrix. To stimulate the potential of the DAG-TDTP model in reducing energy consumption, we propose a genetic algorithm based on temporary key nodes (TKNGA) for the proposed model. Through simulation analysis, we verify the superiority of the proposed model in reducing UAV system energy consumption and the superiority of TKNGA compared to other algorithms.

Memory-processor co-scheduling of AECR-DAG real-time tasks on partitioned multicore platforms with scratchpads

Multicore systems with core-level scratchpad memories offer appealing architectures for constructing efficient and predictable real-time systems. In this work, we aim to improve the usability of scratchpad memories and exploit their predictability to hide access latency to shared resources. We use a genetic algorithm to derive scheduling parameters for a set of directed acyclic task-graphs (DAGs). DAGs consist of dependent subtasks and their respective communications, following the Acquisition Execution Restitution (AER) model. Subtasks are partitioned onto the multicore platform while scheduling their memory requests and relative communications onto the shared buses, in order to prevent interference and ensure predictability. Specifically, all subtasks and communications are assigned appropriate intermediate offsets and deadlines to guarantee that they comply with the system’s timing constraints. We conducted a large set of synthetic experiments to demonstrate the effectiveness of the proposed technique.

Task Offloading of Intelligent Building Based on Dependency-Aware in Edge Computing

Background: With the rapid development of artificial intelligence, the traditional cloud computing model has serious bandwidth and energy consumption problems when storing and processing massive amounts of raw data. To address this problem, recent patents have investigated methods for intelligent task offloading and allocation in mobile edge computing. Objective: A Directed Acyclic Graph (DAG) task unloading model is established to reduce the problem of task delay and energy consumption in edge networks. At the same time, the Modified Tuna Swarm Optimization (MTSO) is used to improve the execution efficiency. Methods: Firstly, this paper integrates (i) inter-task dependencies; (ii) heterogeneity of computing resources in the edge network; and (iii) interference of wireless channels in the edge network. A DAG task offloading model is developed to reduce latency and energy consumption. The end users are guided to offload their tasks/sub-tasks to the most appropriate servers in the edge network, thus minimizing the end-to-end latency of all tasks in the edge network. Secondly, the MTSO algorithm is used to coordinate the dependencies and priorities of subtasks to improve execution efficiency. Results: The experimental results show that when the number of users including subtasks is 10, the final edge server utilization rate is as high as 92%. A more fine-grained segmentation scheme can reduce the average delay of tasks and improve the utilization rate of edge servers. Conclusion: The approach proposed in this paper reduces the end-to-end latency and improves resource utilization in complex applications under the premise of ensuring task dependency. It well relieves the pressure on the cloud and has certain engineering application value

A New Federated Scheduling Algorithm for Arbitrary-Deadline DAG Tasks

A parallel task can always be modelled as a directed acyclic graph (DAG), where sequential instruction blocks are modelled as vertices and data dependencies or resource constraints are modelled as edges. We propose a new federated scheduling algorithm for arbitrary-deadline sporadic DAG tasks, assuming that the exact structures of DAG tasks are unknown before runtime. Federated scheduling algorithms are a class of algorithms that can efficiently schedule DAG tasks by assigning several processors exclusively to each task. Existing studies have shown the advantages of federated scheduling, which include increasing the analytical schedulability and minimising the scheduling overhead. We are particularly focused on the scheduling of any task with a deadline longer than its release period; in this case, multiple jobs generated by the task could run concurrently. For such tasks, our algorithm is different from most federated scheduling algorithms in that it assigns dedicated processors to each job instead of letting jobs released by the same task share processors. The main idea is to increase the analytical schedulability by avoiding interference between jobs. The simulation results show that our algorithm outperforms existing algorithms when the exact structures of tasks are unknown before runtime.

Joint UAV Trajectory Planning, DAG Task Scheduling, and Service Function Deployment Based on DRL in UAV-Empowered Edge Computing

Unmanned Aerial Vehicle (UAV)-empowered edge computing has been widely investigated in obstacle-free scenarios, where a moving UAV is in charge of handling offloaded singleton tasks from mobile devices on the ground. However, little attention has been paid to the scenario, in which the UAV serves a complex area with multiple obstacles and dependent tasks. A dependent task can be formulated as a Directed Acyclic Graph (DAG) that contains a number of sub-tasks; and each sub-task can be executed by a corresponding Service Function (SF) deployed on the UAV. In this backdrop, the joint UAV trajectory planning, DAG task scheduling, and SF deployment is formulated as an optimization problem in this paper. Afterwards, a Deep Reinforcement Learning (DRL)-based algorithm is presented to tackle the NP-hard problem. The state space, action space, and the reward function of the agent, i.e., the UAV, are defined respectively under the DRL framework. To evaluate the effectiveness of the proposal, a series of experiments is conducted with different parameter settings. Results show that the DRL-based algorithm performs much better than three heuristic algorithms in success rate of trajectory planning, the number of executed tasks, and the average task response latency.

Online energy-efficient scheduling of DAG tasks on heterogeneous embedded platforms

Heterogeneous distributed computing systems are becoming more and more common to process practical applications in many embedded systems, where energy efficiency is very important, especially for battery-powered systems. This paper studies the problem of how to adaptively schedule dynamic applications on the fly on a heterogeneous embedded system such that the system’s energy consumption can be minimized, while meeting applications’ response time and reliability requirements simultaneously. We first study the problem of scheduling a set of static applications. This problem is formulated as an integer nonlinear programming problem by deriving the response time equation of directed acyclic graph (DAG) applications. A reliability-aware scheduling strategy is designed to solve this problem. On top of this strategy, we develop an algorithm that tries to use as few processors as possible to finish the computation because, in this way, the system’s static energy can be saved. We then extend this solution to handle the problem of scheduling dynamic applications. By developing a task migration algorithm and a schedule adjustment algorithm, the system can adaptively handle suspended and newly released applications on the fly. Extensive simulation experiments are conducted, and the results demonstrate the high efficiency of our proposed approaches against some classic heuristics and a state-of-the-art approach.

SAFLA: Scheduling Multiple Real-Time Periodic Task Graphs on Heterogeneous Systems

Many modern Cyber Physical Systems (CPSs) are composed of multiple independent periodically executing real-time control tasks having inter-dependent component sub-tasks. Each such control task is therefore usually represented as <i>Directed-acyclic Task Graphs</i> (DTGs). These CPSs are often distributed in nature and are quickly shifting from homogeneous to heterogeneous processing platforms in order to meet ever increasing demands for performance and energy savings, within limited resource budgets. In spite of the practical relevance of the problem in today's CPS design scenario, very few research works in literature have tried to address this due to its inherent computational as well as design complexity. This work endeavors to solve the problem of co-scheduling a set of periodic real-time applications each modelled as an independent DTG, to be executed on a distributed platform consisting of heterogeneous processors communicating using shared buses. Assuming the processing platform to be DVFS (Dynamic Voltage Frequency Scaling) enabled, we attempt to minimize dynamic energy dissipation associated with the execution of all DTGs over an hyperperiod <inline-formula><tex-math notation="LaTeX">$\mathcal {H}$</tex-math></inline-formula> while ensuring that no DTG instance within <inline-formula><tex-math notation="LaTeX">$\mathcal {H}$</tex-math></inline-formula> misses its deadline. The problem has first been formally represented as a constraint optimization problem. However, an optimal solution using standard solvers become prohibitively compute as well as memory intensive and doesn't scale even for moderate problem sizes. Hence, in this work, we attempt to develop a three-phase list-based hierarchical scheduling algorithm called <i>Slack Aware Frequency Level Allocator</i> ( <i>SAFLA</i> ). The efficacy of <i>SAFLA</i> has been critically evaluated through simulation using benchmark DTGs.

Multiobjective Approach to Schedule DAG Tasks on Voltage Frequency Islands

Scheduling a Directed Acyclic Graph (DAG) on voltage frequency islands involves dividing the available processing units into multiple islands with varying voltage and frequency levels, and then mapping the tasks of the DAG to the islands while minimizing the makespan and overall energy consumption. In this research work, a novel DAG task scheduling model is introduced with the assistance acquired from the deep learning paradigm. The proposed model includes four major phases: (a) DAG modelling, (b) Voltage frequency island partitioning, (c) core temperature prediction and (d) scheduling optimization. Initially, the Directed Acyclic Graph (DAG) model is designed. The nodes of DAG represent tasks and the edges represent dependencies between tasks. Then, in the Voltage frequency island partitioning, the available processing units into multiple voltage frequency islands. This can be done based on the power consumption of each unit, and the task requirements. Subsequently, the Recurrent Neural Network (RERNN) is trained to predict the core temperature of each voltage frequency island based on the multi-objectives like execution time, makespan, overall energy consumption. Then, the scheduling of the DAG on the voltage frequency islands is optimized using the Self-Improved Pelican Optimization Algorithm (SI-POA). The proposed SI-POA model is an extended version of the standard POA model. The SI-POA model is inspired by the behavior by the natural behavior of pelicans during hunting. In the scheduling optimization phase, the SI-POA algorithm to optimize the scheduling of the DAG on the voltage frequency islands while taking into account the predicted core temperature of each island based on the Recalling-enhanced recurrent neural network (RERNN) model. The goal is to minimize the makespan and overall energy consumption of the DAG while keeping the core temperature of each island within safe limits.

GMOM: An Offloading Method of Dependent Tasks Based on Deep Reinforcement Learning

Mobile edge computing (MEC) is considered as an effective solution to delay-sensitive services, and computing offloading, the central technology in MEC, can expand the capacity of resource-constrained mobile terminals (MTs). However, because of the interdependency among applications, and the dynamically changing and complex nature of the MEC environment, offloading decision making turns out to be an NP-hard problem. In the present work, a graph mapping offloading model (GMOM) based on deep reinforcement learning (DRL) is proposed to address the offloading problem of dependent tasks in MEC. Specifically, the MT application is first modeled into a directed acyclic graph (DAG), which is called a DAG task. Then, the DAG task is transformed into a subtask sequence vector according to the predefined order of priorities to facilitate processing. Finally, the sequence vector is input into an encoding-decoding framework based on the attention mechanism to obtain the offloading strategy vector. The GMOM is trained using the advanced proximal policy optimization (PPO) algorithm to minimize the comprehensive cost function including delay and energy consumption. Experiments show that the proposed model has good decision-making performance, with verified effectiveness in convergence, delay, and energy consumption.

Toward Minimum WCRT Bound for DAG Tasks Under Prioritized List Scheduling Algorithms

Many modern real-time parallel applications can be modeled as a directed acyclic graph (DAG) task. Recent studies show that the worst-case response time (WCRT) bound of a DAG task can be significantly reduced when the execution order of the vertices is determined by the priority assigned to each vertex of the DAG. How to obtain the optimal vertex priority assignment, and how far from the best-known WCRT bound of a DAG task to the minimum WCRT bound are still open problems. In this article, we aim to construct the optimal vertex priority assignment and derive the minimum WCRT bound for the DAG task. We encode the priority assignment problem into an integer linear programming (ILP) formulation. To solve the ILP model efficiently, we do not involve all variables or constraints. Instead, we solve the ILP model iteratively, i.e., we initially solve the ILP model with only a few primary variables and constraints, and then at each iteration, we increment the ILP model with the variables and constraints which are more likely to derive the optimal priority assignment. Experimental work shows that our method is capable of solving the ILP model optimally without involving too many variables or constraints, e.g., for instances with 50 vertices, we find the optimal priority assignment by involving 12.67% variables on average and within several minutes on average.

Response-Time Analysis of Limited-Preemptive Sporadic DAG Tasks

Guaranteeing timing constraints for parallel real-time applications deployed on multicore platforms is challenging, especially for applications containing non-preemptive execution blocks, that suffer from priority inversions. In this article, we propose to model such applications using a sporadic directed acyclic graph (DAG) model where preemption may take place only between the nodes of a DAG task. We present a new method for response-time analysis of such tasks scheduled with the global fixed-priority scheduling policy. We show that our method outperforms the state-of-the-art techniques significantly in terms of resource utilization in experimental evaluations using both benchmark and randomly generated task sets. We also present a method to deal with global EDF scheduling, which is a new technique proposed for response time analysis of sporadic DAG tasks with non-preemptive nodes.

Improving Interference Analysis for Real-Time DAG Tasks Under Partitioned Scheduling

Real-time systems with strict timing constraints have been widely applied in many fields. The Directed acyclic graph (DAG) task model has been widely studied and applied to model real-time systems with partial parallelism and precedence constraints in each task. Our paper focuses on the worst-case response time (WCRT) analysis of DAG tasks under partitioned scheduling on multiprocessors. We investigate a parallel structure named <formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex>$Str$</tex> </formula> , which helps obtain more accurate analysis results, and propose a new offline scheduling analysis algorithm named reducing repetitive calculation (RRC). Experiments with synthetic workload are conducted to compare the results calculated by RRC and the state-of-the-art, as well as the observed average response time on a real embedded system. Results show that RRC has better performance in terms of analysis accuracy.

Response time analysis of parallel tasks on accelerator-based heterogeneous platforms

Due to the inherent high parallelism and heterogeneity, accelerator-based heterogeneous platforms have been regarded as promising solutions for computation-intensive applications. With respect to latency and power consumption, accelerator often performs better than general-purpose processors. The directed acyclic graph (DAG) model is widely used to represent parallel applications, where each vertex represents an independent workload and each edge represents the precedence constraint between two vertices. The response time analysis of DAG tasks is of paramount importance to the judgment of task schedulability. However, the existing response time analysis method of DAG tasks on accelerator-based heterogeneous platforms just offloading one vertex to the accelerator cannot comply with the actual application scenarios. The well-known worst-case response time (WCRT) bounds of DAG tasks on heterogeneous platforms are pessimistic for judging the task schedulability. This paper studies the response time analysis method of DAG tasks that offload multiple vertices to the accelerator. We transform the graph structure of the DAG task to reduce the interference of vertices running on the accelerator to vertices on the general-purpose processor. Finally, we propose a new WCRT bound for the transformed DAG. The experimental results show that the new WCRT bound is more precise than the existing bounds of DAG tasks.

Adaptive search strategy based chemical reaction optimization scheme for task scheduling in discrete multiphysical coupling applications

Parallel computing problems of multiphysical coupling applications based on discrete grids can be equivalently transformed into parallel computing problems based on a directed acyclic graph (DAG). Due to the problem of the discrete high-dimensional grids of the multiphysical coupling application, the directed data dependencies usually have parallelism. Moreover, this kind of scheduling problem based on DAG is NP-hard problem. Heuristic algorithms are often used to achieve optimal execution order scheduling of tasks and mapping between tasks and processors. In this paper, we propose an improved chemical reaction optimization algorithm based on adaptive search strategy (ASSCRO), which is used to solve the DAG task scheduling problem of discrete multiphysical coupling applications. ASSCRO is divided into two phases. The first phase is used to search the directed execution order of the tasks, and the second phase aims to use the heuristic strategy to map the tasks to the processors efficiently. In the four basic reactions of CRO, the algorithm can cover a larger search solution space by using an adaptive search strategy, it can obtain a better solution, and achieve less overhead and superior performance than the state-of-the-art. We conducted the experiment that applied our ASSCRO to deal with the multiphysical coupling applications. The experimental results showed that the proposed algorithm outperforms other algorithms in multiple metrics when dealing with DAG scheduling problems.

Global Fixed-Priority Scheduling for Parallel Real-Time Tasks with Constrained Parallelism

With the rapid development of parallel programming techniques and the widespread use of multiprocessors, scheduling and analysis techniques supporting parallel real-time tasks become a critical topic for multiprocessor real-time systems. Global scheduling that allows the vertices of a parallel task to execute on any processor is a promising scheduling approach with guaranteed theoretical bounds and wide use in practice. However, the complex internal structure of parallel tasks leads to extensive inner- and inter-task interference, which leads to significant pessimism in the worst-case timing analysis. In this paper, a global fixed-priority (G-FP) scheduling with constrained parallelism for parallel real-time tasks based on the sporadic directed acyclic graph (DAG) model is proposed. Each DAG task is assigned a parallel threshold, such that the number of processors occupied by the task is limited to the parallel threshold of the task at a time. We propose a heuristic algorithm to set the parallel threshold and present a response-time analysis (RTA) to exploit the feature of the constrained parallel scheduling. Experiments with randomly generated tasks show that the proposed approach improves the schedulability upon G-FP and federated scheduling in terms of acceptance ratio.

Computing exact WCRT for typed DAG tasks on heterogeneous multi-core processors

Heterogeneous multi-core architectures achieve high performance and energy efficiency in real-time domain towards various applications. Most real-time parallel applications on heterogeneous multi-cores can be modeled as a typed directed acyclic graph (DAG) task, where the workload of each vertex is only allowed to execute on a particular type of core. Traditional worst-case response time (WCRT) analysis methods for these applications modeled as DAG are very pessimistic and impractical, i.e., there is a big gap between existing WCRT bounds and the exact WCRT. In this paper, we propose a satisfiability modulo theories-based method to exactly analyze the WCRT of the typed DAG task scheduled upon heterogeneous multi-cores. Experimental results show that our method can significantly improve the precision of the WCRT, and thus, dramatically increase the acceptance rate in the schedulability analysis.

Worst-Case Response Time Analysis of Multitype DAG Tasks Based on Reconstruction

With the wide application of heterogeneous multi-core processor real-time systems, the existing analysis methods of worst-case response time (WCRT) overestimate the blocking information among tasks, resulting in a rather pessimistic estimation. To improve the accuracy of the WCRT, we propose a reconstruction-based WCRT analysis method for multi-type directed acyclic graph (DAG) tasks scheduling algorithm(RMDS). The RMDS algorithm comprises the following steps: First, we unitize all task nodes in the multi-type DAG task; Then, we use key factors as task priorities to schedule tasks and reconstruct the DAG task model into a parallel node segment model; Finally, we estimate the WCRT of multi-type DAG tasks according to the parallel node segment model to assess task schedulability. To verify the performance of our algorithm, we compared it with traditional algorithms. RDMS showed an acceptance rate 6.13% higher and its overall performance increased by 25.95% in comparison with traditional algorithms.

Decomposition-based scheduling for parallel real-time tasks on multiprocessors

This paper addresses the problem of multiprocessor real-time scheduling for parallel tasks modeled as directed acyclic graph (DAG). We propose a new decomposition-based scheduling algorithm to schedule parallel DAG tasks with implicit-deadlines. We prove that our algorithm has a resource augmentation bound of 2.618 for DAG tasks with implicit-deadlines when the decomposed tasks are scheduled using Global EDF scheduling, i.e., if the input set of DAG tasks is schedulable on m unit-speed processors, the proposed scheduling algorithm will always succeed on m processors with speed 2.618. This result improves the resource augmentation bound of 2.618 for tasks with special DAGs provided by Qamhieh et al. in 2019. Moreover, when using Global DM (Partitioned-DM, respectively) scheduling for the decomposed tasks, we prove a resource augmentation bound of 3.42 for DAG tasks with implicit-deadlines, which improves the resource augmentation bound of 3.73 for tasks with special DAGs (3.42 for fork-join tasks, respectively).

Failure-resilient DAG task scheduling in edge computing

Through placing computation, storage, and communications facilities near the data source, Edge Computing (EC) is anticipated to extend the intelligence from the central cloud to the network edge. The Quality of Experience (QoE) of user and energy efficiency of mobile device could be significantly improved through offloading their computation-intensive tasks to the network edge. With the increasing popularity of intelligent devices, tasks offloaded to the edge are becoming more complex, consisting of multiple sub-tasks with data dependency, which are typically modeled as a Directed Acyclic Graph (DAG). The scheduling of DAG tasks is more complex, which has been proved to be NP-hard. Traditional DAG scheduling algorithms developed in non-edge computing scenarios could not be directly applied due to their neglect of: (1) the competition of communication resources; and (2) the rescheduling requirement in case of edge server failure in dynamic edge network environment. In this backdrop, this paper presents a failure-resilient DAG task scheduling algorithm to minimize the response delay experienced by the tasks. After formulating the DAG task scheduling problem, a context-aware greedy task scheduling (CaGTS) algorithm is proposed. Then, to cope with the failure event of edge server, a dependency-aware task rescheduling (DaTR) algorithm is designed. To evaluate the performance of the proposed algorithms, extensive experiments have been conducted on a simulator developed using Python. Experimental results with diverse parameter settings have shown that CaGTS could reduce at least 10.47% average completion time than benchmarks, and DaTR can effectively avoid task scheduling interruption caused by server failure events.

Contention Cognizant Scheduling of Task Graphs on Shared Bus-Based Heterogeneous Platforms

Demands for high performance as well as reliability within stringent resource budgets are driving a shift from homogeneous to heterogeneous processing platforms for the implementation of today&#x2019;s cyber-physical systems (CPSs). These CPSs are typically represented as directed-acyclic task graphs (DTGs) due to the complex interactions between their functional components which are often distributed in nature. This work deals with the problem of scheduling a CPS modeled as DTG. First, we present an optimal solution using integer linear programming (ILP) for the DTGs, to be executed on distributed heterogeneous processors which are interconnected via shared buses. However, this ILP-based optimal solution exhibits high computational complexity and does not scale for moderately large problem sizes. Hence, we propose a low-overhead heuristic algorithm called the contention cognizant task and message scheduler (CC-TMS), which is able to produce satisfactorily efficient as well as fast solutions within a reasonable time. The efficiency of the proposed scheme has been extensively evaluated through simulation-based experiments using benchmark DTGs. Through the case study of a real-world <i>automotive traction controller</i>, we demonstrate the practical applicability of our proposed scheme.