Heterogeneous Multiprocessor Systems Research Articles

A hybrid optimization algorithm for energy-aware multi-objective task scheduling in heterogeneous multiprocessor systems

A hybrid optimization algorithm for energy-aware multi-objective task scheduling in heterogeneous multiprocessor systems

METSM: Multiobjective energy-efficient task scheduling model for an edge heterogeneous multiprocessor system

Along with the growth of computing requirements, edge devices, such as satellites and unmanned aerial vehicles, are equipped with heterogeneous multiprocessor systems to cope with complicated missions, including attitude control, signal processing, and objective detection. Concerning the extremely limited power on these devices, reducing the energy and time overhead of task execution evolves into a crucial challenge. Therefore, a multiobjective energy-efficient task scheduling technique (METSM) is proposed. First, a mathematical model is established for the energy-efficient task scheduling problem on edge heterogeneous multiprocessor systems. In this model, both makespan and total energy consumption are the optimization objectives. The decision variables include the task execution sequence, processor assignment, and dynamic voltage and frequency scaling level for each processor. Second, a problem-specific algorithm, namely, iterated greedy-based multiobjective optimizer (IMO), is proposed. Specifically, destruction–reconstruction and local search are redesigned for optimizing task scheduling and resource allocation. Considering local optima avoidance, a probabilistic mutation operation is developed. In addition, multiobjective-oriented strategies of optimal solution selection and acceptance criteria are adopted to accelerate the convergence. Finally, our proposed IMO is compared with several of the latest algorithms through multiple performance metrics. The experimental results show that IMO can obtain optimal Pareto fronts among several multiobjective methods. Savings of approximately 10% and 12% in time and power consumption, respectively, can be achieved by IMO. Moreover, in comparison with classic list-based heuristics for solving test cases, while IMO maintains a similar makespan, energy is reduced by nearly 90%.

MaGNAS: A Mapping-Aware Graph Neural Architecture Search Framework for Heterogeneous MPSoC Deployment

Graph Neural Networks (GNNs) are becoming increasingly popular for vision-based applications due to their intrinsic capacity in modeling structural and contextual relations between various parts of an image frame. On another front, the rising popularity of deep vision-based applications at the edge has been facilitated by the recent advancements in heterogeneous multi-processor Systems on Chips (MPSoCs) that enable inference under real-time, stringent execution requirements. By extension, GNNs employed for vision-based applications must adhere to the same execution requirements. Yet contrary to typical deep neural networks, the irregular flow of graph learning operations poses a challenge to running GNNs on such heterogeneous MPSoC platforms. In this paper, we propose a novel unified design-mapping approach for efficient processing of vision GNN workloads on heterogeneous MPSoC platforms. Particularly, we develop MaGNAS, a mapping-aware Graph Neural Architecture Search framework. MaGNAS proposes a GNN architectural design space coupled with prospective mapping options on a heterogeneous SoC to identify model architectures that maximize on-device resource efficiency. To achieve this, MaGNAS employs a two-tier evolutionary search to identify optimal GNNs and mapping pairings that yield the best performance trade-offs. Through designing a supernet derived from the recent Vision GNN (ViG) architecture, we conducted experiments on four (04) state-of-the-art vision datasets using both ( i ) a real hardware SoC platform (NVIDIA Xavier AGX) and ( ii ) a performance/cost model simulator for DNN accelerators. Our experimental results demonstrate that MaGNAS is able to provide 1.57 × latency speedup and is 3.38 × more energy-efficient for several vision datasets executed on the Xavier MPSoC vs. the GPU-only deployment while sustaining an average 0.11% accuracy reduction from the baseline.

Optimization Task Scheduling Using Cooperation Search Algorithm for Heterogeneous Cloud Computing Systems

Cloud computing has taken over the high-performance distributed computing area, and it currently provides on-demand services and resource polling over the web. As a result of constantly changing user service demand, the task scheduling problem has emerged as a critical analytical topic in cloud computing. The primary goal of scheduling tasks is to distribute tasks to available processors to construct the shortest possible schedule without breaching precedence restrictions. Assignments and schedules of tasks substantially influence system operation in a heterogeneous multiprocessor system. The diverse processes inside the heuristic-based task scheduling method will result in varying makespan in the heterogeneous computing system. As a result, an intelligent scheduling algorithm should efficiently determine the priority of every subtask based on the resources necessary to lower the makespan. This research introduced a novel efficient scheduling task method in cloud computing systems based on the cooperation search algorithm to tackle an essential task and schedule a heterogeneous cloud computing problem. The basic idea of this method is to use the advantages of meta-heuristic algorithms to get the optimal solution. We assess our algorithm’s performance by running it through three scenarios with varying numbers of tasks. The findings demonstrate that the suggested technique beats existing methods New Genetic Algorithm (NGA), Genetic Algorithm (GA), Whale Optimization Algorithm (WOA), Gravitational Search Algorithm (GSA), and Hybrid Heuristic and Genetic (HHG) by 7.9%, 2.1%, 8.8%, 7.7%, 3.4% respectively according to makespan.

Task Scheduling Optimization in Cloud Computing by Coronavirus Herd Immunity Optimizer Algorithm

Cloud computing is now dominant in high-performance distributed computing, offering resource polling and ondemand services over the web. So, the task scheduling problem in a cloud computing environment becomes a significant analysis space due to the dynamic demand for user services. The primary goal of scheduling tasks is to allocate tasks to processors to achieve the shortest possible makespan while respecting priority restrictions. In heterogeneous multiprocessor systems, task and schedule assignments significantly impact the system's operation. Therefore, the different processes within the heuristic-based scheduling task algorithm will lead to a different makespan on a heterogeneous computing system. Thus, a suitable algorithm for scheduling should set precedence efficiently for every subtask based on the resources required to reduce its makespan. This paper proposes a novel efficient scheduling task algorithm based on the coronavirus herd immunity optimizer algorithm to solve task scheduling problems in a cloud computing environment. The basic idea of this method is to use the advantages of meta-heuristic algorithms to get the optimal solution. We evaluate the performance of our algorithm by applying it to three cases. The collected findings suggest that the proposed strategy successfully achieved the best solution in terms of makespan, speedup, efficiency, and throughput compared to others. Furthermore, the results demonstrate that the suggested technique beats existing methods new genetic algorithm (NGA), proposed particle swarm optimization (PPSO), whale optimization algorithm (WOA), enhanced genetic algorithm for task scheduling (EGA-TS), gravitational search algorithm (GSA), genetic algorithm (GA), and hybrid heuristic and genetic (HHG) by 22.8%, 12.3%, 8.8%, 7.3%, 7.3%, 3.4%, and 3.4% respectively according to makespan.

A novel algorithm for priority-based task scheduling on a multiprocessor heterogeneous system

Task scheduling is a major challenging issue in parallel and distributed heterogeneous computing systems. The task scheduling problem in a heterogeneous multiprocessor system typically comprises two phases: assigning priority to tasks and mapping tasks to processors. This work considers the task scheduling problem a multi-objective optimization problem to minimize the makespan and maximize the processor utilization. In this paper, the authors have proposed a novel dynamic task scheduling algorithm that is based on an existing metaheuristic algorithm called Black Widow Optimization (BWO) and an existing deep learning method called Deep Recurrent Neural Network (DRNN) to achieve the objective. The scheduling algorithm is called DRNN-BWO (DRNN trained by using BWO). Authors have used the Upward Rank technique to prioritize the tasks and the Priority-based Modules Encoding (PME) technique for mapping tasks to processors. The simulation results of the proposed algorithm are compared with some standard algorithms to find out how they fare against each other in reducing the makespan. The proposed algorithm is also compared with some recently developed metaheuristic algorithms based on some real-life applications. All the simulation results show that the proposed algorithm significantly reduces the makespan and increases processor utilization.

Time‐constrained and reliability aware energy minimization scheduling algorithm for heterogeneous multiprocessor environments

SummaryHeterogeneous multiprocessor systems are now widely used in industry by providing high performance and high concurrency. However, with the increasing number of computational nodes, it leads to a dramatic increase in energy consumption of heterogeneous multiprocessor systems. Most of the current research has been conducted to reduce the energy consumption by reducing the processor frequency and extending the task execution time, but these measures often lead to a significant decrease in system reliability. This article addresses the problem of energy‐aware task scheduling in the case of distributed computing systems with deadline and reliability constraints. First, a time and reliability allocation model is built to assign the deadline and reliability constraints to each task, which ensures fairness among tasks. Then, a three‐stage scheduling algorithm is proposed to minimize the system energy consumption. The static energy consumption is minimized by shutting down the inefficient processors. The dynamic energy consumption of the system is reduced by distributing tasks as evenly as possible on each processor through a task redistribution strategy. Finally, experimental results on real scientific workflows and randomly generated graphs show that the proposed algorithm outperforms other algorithms in terms of energy reduction.

Energy-aware scheduling for dependent tasks in heterogeneous multiprocessor systems

Heterogeneous multiprocessor platform has been widely adopted as an effective approach to providing strong calculation capability while keeping complexity and energy consumption under control in large-scale systems. Although this platform is able to achieve efficient cost reduction and flexibility enhancement in the design and development process of real-time applications, it brings a serious and complex multi-task scheduling problem, especially for dependent tasks with energy consumption constraints. All tasks should be scheduled according to appropriate strategies such that their dependence requirements and energy consumption limitations would be satisfied even in the worst-case situations. In this work, we focus on the energy-aware scheduling problem of dependent tasks in heterogeneous multiprocessor systems. First, we model the dependent tasks and heterogeneous processors, and formulate the energy-aware scheduling problem as a constrained optimization one with an objective of minimizing the schedule length of tasks. Then, by adopting an efficient task prioritization strategy and a weight-based energy distribution strategy, we propose a list-based energy-aware scheduling algorithm to seek an approximate optimal start time and processor allocation for each task, guaranteeing that all tasks would be executed efficiently while meeting the dependence and energy requirements. Experiments with randomly generated tasks are conducted to evaluate the performances of the proposed approach in terms of schedule length, optimal solution ratio, and execution time.

The problem of developing the architecture of modern cognitive radar system

The problem of developing the architecture of modern cognitive radar systems using artificial intelligence technologies is considered. The main difference from traditional systems is the use of a trained neural network. The heterogeneous multiprocessor system is rebuilt in the process of solving the problem, providing reliability and solving various types of problems of one class and deep learning of the neural network in real time. This architecture promotes the introduction of cognitive technologies that take into account the requirements for the purpose, the influence of external and internal factors.

Task Scheduling Optimization in Cloud Computing by Rao Algorithm

Cloud computing is currently dominated within the space of high-performance distributed computing and it provides resource polling and on-demand services through the web. So, task scheduling problem becomes a very important analysis space within the field of a cloud computing environment as a result of user's services demand modification dynamically. The main purpose of task scheduling is to assign tasks to available processors to produce minimum schedule length without violating precedence restrictions. In heterogeneous multiprocessor systems, task assignments and schedules have a significant impact on system operation. Within the heuristic-based task scheduling algorithm, the different processes will lead to a different task execution time (makespan) on a heterogeneous computing system. Thus, a good scheduling algorithm should be able to set precedence efficiently for every subtask depending on the resources required to reduce (makespan). In this paper, we propose a new efficient task scheduling algorithm in cloud computing systems based on RAO algorithm to solve an important task and schedule a heterogeneous multiple processing problem. The basic idea of this process is to exploit the advantages of heuristic-based algorithms to reduce space search and time to get the best solution. We evaluate our algorithm's performance by applying it to three examples with a different number of tasks and processors. The experimental results show that the proposed approach significantly succeeded in finding the optimal solutions than others in terms of the time of task implementation.

TEA-SEA: Tiling and scheduling of non-uniform two-level perfectly nested loops using an evolutionary approach

Nowadays, in scientific and computational programs, increasing the execution speed of programs is very important. In implementing scientific programs, loops are dedicating a large part of time to themselves. Therefore, the parallel execution of the iteration loop reduces the runtime of all programs. This paper has been working on the parallelism of two-level perfectly nested loops with non-uniform dependencies and rectangular iteration space, and then their scheduling is based on the scheduling of task graphs. For this purpose, the iteration space of non-uniform loop is partitioned using a tiling transformation, and then the tiled space is modelled as a task, graph for scheduling. In the task graph, vertices expressive tiles and edges expressive data dependencies inter-tiles. It should be noted that the computational load of vertices is not the same, and the vertices corresponding imperfect tiles have a less computational load. The scheduling is done on both types of homogeneous and heterogeneous multiprocessor systems. The tiling of iteration space loop and the scheduling of task graph is arranged in the category of NP-Hard problems that using evolutionary algorithms can be efficient to discover a suitable solution for them. The results of the proposed approach show the performance of this method better than previous methods in the tiling and scheduling of non-uniform loops. According to the examinations performed on uniform loops, this algorithm also performs better than previous algorithms on these types of loops.

Reliability-aware task scheduling for energy efficiency on heterogeneous multiprocessor systems

Recent studies mainly focus on high performance or low power consumption for task scheduling on heterogeneous multiprocessor systems (HMSs). Dynamic voltage and frequency scaling (DVFS) is an important energy reduction technique, which adjusts the voltage and frequency of the processor while the task is executing. However, some studies have shown that reducing the voltage of processor increases the transient failure rate, which reduces system reliability. In this paper, we aim at addressing the scheduling problem of optimizing energy under makespan and reliability constraints on HMSs with DVFS. We first propose an improved whale optimization algorithm (WOA) deploying opposition-based learning and individual selection strategy, which can balance the exploration and exploitation ability. To maintain population diversity, we then apply a constrained rank-based method which retains some infeasible individuals in the population. Finally, we reschedule the Critical Path Nodes (CPNs) to further improve the performance of improved WOA. The main difference between our work and most previous works is that we study a new scheduling problem, and utilize an improved WOA algorithm integrating with rescheduling CPNs and a constrained rank-based method. Extensive experiments are conducted to evaluate our proposed algorithm, and the evaluation results show that our proposed algorithm is compelling in comparison with the state-of-the-art algorithms.

Task scheduling on heterogeneous multiprocessor systems through coherent data allocation

AbstractEnergy consumption has become one of the main bottlenecks that limit the performance improvement of heterogeneous multiprocessor systems. In a heterogeneous distributed shared‐memory multiprocessor system (HDSMS), each processor can access all the memories, and each data can be stored in different memories. This article aims at addressing the problem of task scheduling and data allocation (TSDA) on HDSMS. To minimize the total energy consumption under a time constraint for TSDA, we propose two algorithms: the extended tree assignment for task scheduling incorporating data allocation (ETATS‐DA) and critical path task scheduling and data allocation (CPTSDA). The ETATS‐DA algorithm first utilizes the extended tree assignment to search the near optimal solution for task assignment, and then allocates data to memory based on the result of assignment. The CPTSDA algorithm considers TSDA jointly on a critical path simultaneously. Our proposed algorithms perform coherent data allocation under the consideration of best task scheduling by running two different heuristic strategies, respectively, and taking the best result as the final result. We conduct a large number of simulation experiments to test the performance of our algorithms, and the results validate the higher performance of our methods compared with the state‐of‐the‐art algorithms.

Security-Critical Energy-Aware Task Scheduling for Heterogeneous Real-Time MPSoCs in IoT

Internet of Things (IoT) devices, such as intelligent road side units and video-based detectors, are being deployed in emerging applications like sustainable and intelligent transportation systems. The primary obstacles against the development of these IoT devices are various security threats and huge energy consumption. In this article, we study the problem of scheduling tasks onto a heterogeneous multiprocessor system on a chip (MPSoC) deployed in IoT for optimizing quality of security under energy, real-time, and task precedence constraints. We first provide a mixed-integer linear programming (MILP) formulation for allocating and scheduling dependent tasks with energy and real-time constraints on a heterogeneous MPSoC system to maximize system quality of security. In order to efficiently solve the formulated MILP, we then propose an analysis-based two-stage scheme that determines the allocation, operating frequency, and security service of tasks to maximize system quality of security while satisfying the design constraints. We finally carry out extensive simulation experiments to validate our proposed two-stage scheme and MILP approach. Simulation results demonstrate that the proposed two-stage scheme outperforms a number of representative existing approaches in saving energy and improving system quality of security. The results also show that the proposed MILP approach can achieve the best performance and the proposed two-stage scheme has a close performance to the MILP approach.

Bayesian optimization algorithm for multi-objective scheduling of time and precedence constrained tasks in heterogeneous multiprocessor systems

This paper presents a Bayesian optimization based novel approach for multi-objective scheduling of time and precedence constrained tasks in heterogeneous multiprocessing environments. The proposed approach, termed as multi-objective Bayesian optimization algorithm (moBOA) for real-time scheduling, can suitably produce optimal task schedules without any violation of the timing and precedence constraints. The moBOA utilizes Bayesian networks to learn the task graphs that represent the precedence relationships among the tasks. It first allocates tasks to individual processors and then decides the order of execution on each processor based on the latest deadline first policy. The proposed moBOA may be applied to both homogeneous and heterogeneous multiprocessor systems. Extensive comparative analysis has been made by considering two other existing evolutionary algorithms, namely the multi-objective genetic algorithm (moGA) and multi-objective hybrid genetic algorithm (mohGA) through experimental simulations with benchmark datasets. From the results of the simulation experiments, it is observed that moBOA outperforms both moGA and mohGA in terms of quality of solutions, Pareto-optimalilty and standard performance measures. We demonstrate that the task schedules produced by moBOA are optimal and comply with the timing as well as the precedence constraints. Statistical significant analyses of the results are conducted.

Reliability Aware Energy Optimized Scheduling of Non-Preemptive Periodic Real-Time Tasks on Heterogeneous Multiprocessor System

Higher reliability and lower energy consumption are conflicting, yet among the most important design objectives for the real-time systems. Moreover, in the domain of real-time systems, non-preemptive scheduling is relatively unexplored with objectives such as reliability and energy. Thus we propose an active replication based framework to schedule a set of periodic real-time tasks in the non-preemptive heterogeneous environment such that the given reliability and timing constraints are satisfied whereas the energy consumption is minimized. First, we formulate the problem as a constraint optimization problem that provides an optimal solution; however, it does not scale well. Thus, we also propose heuristics which apply reservation of processors and reallocation of jobs, to compute suboptimal solution efficiently in terms of energy consumption as well as schedulability. Heuristics make use of the interplay of task-level reliability target, reliability of replicas, number of replicas, reliability of tasks, and energy consumption. We perform an experimental study on the test cases generated by extending UUnisort algorithm [1] and observe the effect of various simulation parameters on energy consumption and schedulability.

Reliability-Based Heuristic-Based Algorithm for Task Execution in Multiprocessor Systems

Objective: This study implements a combination of heuristicbased algorithm and finding variable neighborhoods, thereby reducing make-span and improving reliability. Methods/statistical analysis: The bi-objective algorithm is proposed for a static planning strategy to achieve high performance in heterogeneous multiprocessor systems. The reliability of a system is based on the probability in which resources of the system execute tasks without any failure. Findings: Here, a genetic algorithm integrated using single neighborhood structure Genetic Variable Neighbourhood Search (GVNS) is implemented to improve the efficient search quality. Applications and improvements: Simulation is performed to maintain better performance parameters when compared with conventional algorithms. Keywords: Directed Acyclic Graph, Inter Process Communication, Evolutionary Algorithm.

Energy-Aware Task Scheduling on Heterogeneous Computing Systems With Time Constraint

As a technique to help achieve high performance in parallel and distributed heterogeneous computing systems, task scheduling has attracted considerable interest. In this paper, we propose an effective Cuckoo Search algorithm based on Gaussian random walk and Adaptive discovery probability which combined with a cost-to-time ratio Modification strategy (GACSM), to address task scheduling on heterogeneous multiprocessor systems using Dynamic Voltage and Frequency Scaling (DVFS). First, to overcome the shortcomings of poor performance in exploitation of the cuckoo search algorithm, we use chaos variables to initialize populations to maintain the population diversity, a Gaussian random walk strategy to balance the exploration and exploitation capabilities of the algorithm, and an adaptive discovery probability strategy to improve population diversity. Then, we apply the improved Cuckoo Search (CS) algorithm to assign tasks to resources, and a widely used downward rank heuristic strategy to find the corresponding scheduling sequence. Finally, we apply a cost-to-time ratio improvement strategy to further improve the performance of the improved CS algorithm. Extensive experiments are conducted to evaluate the effectiveness and efficiency of our method. The results validate our approach and show its superiority in comparison with the state-of-the-art methods.

Online Resource Management for Improving Reliability of Real-Time Systems on “Big–Little” Type MPSoCs

Heterogeneous multiprocessor systems on a chips (MPSoCs) consisting of cores with different performance/power characteristics are widely used in many real-time embedded systems, where both soft-error reliability and lifetime reliability are key concerns. Although existing efforts have investigated related problems, they either focus on one of the two reliability concerns or propose time-consuming scheduling algorithms that cannot adequately address runtime workload and environmental variations. This paper introduces an online framework which is adaptive to runtime variations and maximizes soft-error reliability while satisfying the lifetime reliability constraint for soft real-time systems executing on MPSoCs that are composed of high-performance cores and low-power (LP) cores. Based on each core’s executing frequency and utilization, the framework performs workload migration between high-performance cores and LP cores to reduce power consumption and improve soft-error reliability. Experimental results based on different hardware platforms show that the proposed approach reduces the probability of failures due to soft errors by at least 17% and 50% on average compared to a number of representative existing approaches that satisfy the same lifetime reliability constraints.

Numerical Representation of Directed Acyclic Graphs for Efficient Dataflow Embedded Resource Allocation

Stream processing applications running on Heterogeneous Multi-Processor Systems on Chips (HMPSoCs) require efficient resource allocation and management, both at compile-time and at runtime. To cope with modern adaptive applications whose behavior can not be exhaustively predicted at compile-time, runtime managers must be able to take resource allocation decisions on-the-fly, with a minimum overhead on application performance. Resource allocation algorithms often rely on an internal modeling of an application. Directed Acyclic Graph (DAGs) are the most commonly used models for capturing control and data dependencies between tasks. DAGs are notably often used as an intermediate representation for deploying applications modeled with a dataflow Model of Computation (MoC) on HMPSoCs. Building such intermediate representation at runtime for massively parallel applications is costly both in terms of computation and memory overhead. In this paper, an intermediate representation of DAGs for resource allocation is presented. This new representation shows improved performance for run-time analysis of dataflow graphs with less overhead in both computation time and memory footprint. The performances of the proposed representation are evaluated on a set of computer vision and machine learning applications.