HEFT Algorithm Research Articles

Efficient Task Scheduling for Large-scale Graph Data Processing in Cloud Computing: A Particle Swarm Optimization Approach

To fulfil the requirements of task scheduling for processing massive amounts of graph data in cloud computing environments, this thesis offers the LGPPSO method, which is based on Particle Swarm Optimisation. The LGPPSO algorithm considers the task’s overall structure when initialising numerous particles in order to broaden the search range and raise the likelihood of finding an approximation optimal solution. We evaluated it in large-scale simulation trials with 100 performance-heterogeneous virtual machines (VMs) using both randomly generated and real application graphs, and evaluated its effectiveness against the CCSH and HEFT algorithms. The experimental findings demonstrate that, in both randomly generated graphs and real graph structure applications, significantly reducing the scheduling duration of large-scale graph data is the LGPPSO algorithm. For randomly generated 200 and 400 node tasks, respectively, the scheduling length is shortened by approximately 8.3% and 9.7% on average when compared to the HEFT algorithm. The LGPPSO technique minimises the scheduling length for actual graphical structure applications by an average of 14.6% and 16.9% for the Gaussian 100 and 200 node examples, respectively.

Comparison of the Execution Step Strategies on Scheduling Data-intensive Workflows on IaaS Cloud Platforms

The IaaS platforms of the Cloud hold promise for executing parallel applications, particularly data-intensive scientific workflows. An important challenge for users of these platforms executing scientific workflows is to strike the right trade-off between the execution time of the scientific workflow and the cost of using the platform. In a previous article, we proposed an efficient approach that assists the user in finding this compromise. This approach requires an algorithm aimed at minimizing the execution time of the workflow once the platform configuration is set. In this article, we compare two different strategies for executing a workflow after its offline scheduling using an algorithm. The algorithm that we proposed in the previous study has outperform the HEFT algorithm.
 The first strategy allows some ready tasks to execute earlier than other higher-priority tasks that are ready later due to data transfer times. This strategy is justified by the fact that although our scheduling algorithm attempts to minimize data transfers between tasks running on different virtual machines, this algorithm does not include data transfer times in the planned execution dates for the various tasks of the workflow. The second strategy strictly adheres to the predetermined order among tasks scheduled on the same virtual machine.
 The results of our evaluations show that the best execution strategy depends on the characteristics of the workflow. For each evaluated workflow, our results demonstrate that our scheduling algorithm combined with the best execution strategy surpasses HEFT. The choice of the best strategy must be determined experimentally following realistic simulations, such as the ones we conduct here using the WRENCH framework, before conducting simulations to find the best compromise between cost and execution time of a workflow on an IaaS platform..

A Multi-Objective Cloud Workflow Scheduling Strategy Based on the Modified HEFT Algorithm

Aiming at the issues of poor system resource utilization and the challenge of ensuring the quality of service (QoS) in scientific workflow scheduling in the current cloud computing environment. By combining the existing IaaS cloud resource model with the scientific workflow task model, the article proposed a cloud workflow scheduling model framework. Firstly, proposed multi-objective QoS constraints based on the execution time and costs in scientific workflow scheduling. Then, modified the HEFT algorithm which was a typical scientific workflow scheduling algorithm. Based on the modified HEFT algorithm makes full use of the idle time of resources and maximizes the multi-objective QoS requirements of users. The comparative experiments show that the scheduling strategy of the modified HEFT algorithm can effectively optimize the costs and execution time of scientific workflow scheduling and achieve higher system resource utilization.

Workflow simulation and multi-threading aware task scheduling for heterogeneous computing

Efficient application scheduling is critical for achieving high performance in heterogeneous computing systems. This problem has proved to be NP-complete even for the homogeneous case, heading research efforts in obtaining low complexity heuristics that produce good quality schedules. Such an example is HEFT, one of the most efficient list scheduling heuristics in terms of makespan and robustness.In this paper, we propose two task scheduling methods for heterogeneous computing systems that can be integrated to several task scheduling algorithms. First, a method that improves the scheduling time (the time for obtaining the output schedule) of a family of task scheduling algorithms is delivered without sacrificing the schedule length, when the computation costs of the application tasks are unknown. Second, a method that improves the scheduling length (makespan) of several task scheduling algorithms is proposed, by identifying which tasks are going to be executed as single-threaded and which as multi-threaded implementations, as well as the number of the threads used. We showcase both methods by using HEFT popular algorithm, but they can be integrated to other algorithms too, such as HCPT, HPS, PETS and CPOP.The experimental results, which consider 14580 random synthetic graphs and five real world applications, show that by enhancing HEFT algorithm with the two proposed methods, significant makespan gains and high scheduling time gains, are achieved.

The Application of AI Technology in GPU Scheduling Algorithm Optimization

With the rapid development of integrated circuit technology, GPU computing capabilities continue to improve. Due to the continuous improvement and improvement of GPU programming capabilities, functions, and performance, GPUs have been widely used in the field of high-tech general-purpose computers. This article is aimed at studying the optimization of GPU scheduling algorithm based on AI technology. Through a combination of theoretical analysis and simulation experiments, the concepts of artificial intelligence technology and GPU scheduling are explained, and the impact of GPU architecture and GPGPU load on the energy efficiency of GPGPU is explained. On the basis of comprehensive analysis of GPU cluster characteristics, a new GA-TP scheduling algorithm based on genetic algorithm was designed, and based on the energy efficiency of the cluster, a simulation verification platform was built for the accuracy of simulation. Experimental results show that the acceleration rate of the GA-TP algorithm is significantly lower than that of the HEFT algorithm, the average acceleration rate is reduced by nearly 25%, and the scheduling efficiency of the GA-TP algorithm is higher.

Traffic Routing-Based Computation Offloading in Cybertwin-Driven Internet of Vehicles for V2X Applications

With the rapid development of the Internet of Vehicles (IoV), a lot of Vehicle-to-Everything (V2X) applications have sprung up. To tackle the conflict between the resource-hungry V2X applications and the resource-constrained vehicles, most works focus on the computation offloading problem, which is significant to V2X applications by bringing computation tasks from the vehicles to the edge or cloud infrastructure. However, the dynamic network conditions caused by the mobility of vehicles will bring task migration and huge additional costs, resulting in poor latency performance. Motivated by the aforementioned problem, a traffic routing-based computation offloading scheme in cybertwin-driven IoV for V2X applications is proposed, in which cybertwin represents the network hardware devices and the network software functions. Moreover, according to the cybertwin-driven IoV network architecture, the traffic routing-based computation offloading problem is formulated. Finally, to avoid the inconsistency between the data transmission direction and the vehicle’s movement direction, the enhanced Heterogeneous Earliest Finish Time (eHEFT) algorithm is designed, which introduces the gradient routing into the traditional HEFT algorithm. Performance evaluation results validate that the proposed joint optimization scheme is indeed capable of reducing latency.

MOEAGAC: an energy aware model with genetic algorithm for efficient scheduling in cloud computing

PurposeMajority of the research work either concentrated on the optimization of scheduling length and execution cost or energy optimization mechanism. This research aims to propose the optimization of makespan, energy consumption and data transfer time (DTT) by considering the priority tasks. The research work is concentrated on the multi-objective approach based on the genetic algorithm (GA) and energy aware model to increase the efficiency of the task scheduling.Design/methodology/approachCloud computing is the recent advancement of the distributed and cluster computing. Cloud computing offers different services to the clients based on their requirements, and it works on the environment of virtualization. Cloud environment contains the number of data centers which are distributed geographically. Major challenges faced by the cloud environment are energy consumption of the data centers. Proper scheduling mechanism is needed to allocate the tasks to the virtual machines which help in reducing the makespan. This paper concentrated on the minimizing the consumption of energy as well as makespan value by introducing the hybrid algorithm called as multi-objective energy aware genetic algorithm. This algorithm employs the scheduling mechanism by considering the energy consumption of the CPU in the virtual machines. The energy model is developed for picking the task based on the fitness function. The simulation results show the performance of the multi-objective model with respect to makespan, DTT and energy consumption.FindingsThe energy aware model computes the energy based on the voltage and frequency distribution to the CPUs in the virtual machine. The directed acyclic graph is used to represent the task dependencies. The proposed model recorded 5% less makespan compared against the MODPSO and 0.7% less compared against the HEFT algorithms. The proposed model recorded 125 joules energy consumption for 50 VMs when all are in active state.Originality/valueThis paper proposed the multi-objective model based on bio-inspired approach called as genetic algorithm. The GA is combined with the energy aware model for optimizing the consumption of the energy in cloud computing. The GA used priority model for selecting the initial population and used the roulette wheel selection method for parent selection. The energy model is used as fitness function to the GA for selecting the tasks to perform the scheduling.

Energy‐aware autoscaling for scientific workflow in cloud environment

SummaryThe devices used for information and communication purpose consume an increasing amount of energy. As a result, the datacenters consume a lot of energy. Due to the expanding use of cloud services, there is considerable interest in reducing the energy consumption of datacenters. Effective resource utilization techniques and reduction in energy consumption are two crucial metrics in today's cloud computing environment. Many issues come along with contradicting objectives, viz., minimizing energy consumption raising the expenses of service supply. Autoscaling features of the cloud can bundle these issues. The major hurdle for effective resource scalability is the lack of proper management of energy supply to devices of servers in the datacenter. The main objective of the article is to supply energy to devices as per the allocated job‐size on the interaction of complex energy, to increase the energy utilization of servers. Also, using renewable energy (RE) to supplement the operational energy of the cloud datacenter. In this direction, the article presents a predator‐prey‐based mathematical model for energy scaling that leverages the accessibility of RE source to decrease the high energy demands of the datacenter. It is formalized using ecological concepts of metabolism and allometric scaling. An allometric energy scaling (AES) algorithm is presented. The performance and evaluation of the proposed algorithm are implemented with the Cloudsim 4.0 simulator. The results are compared with the HEFT algorithm, and AES algorithm gives slightly better results in the case of optimum energy consumption and processing cost compared with the HEFT algorithm.

An HC-CSO Algorithm for Workflow Scheduling in Heterogeneous Cloud Computing System

Many scientists are using meta-heuristic techniques for dynamic workflow task scheduling in the area of cloud computing systems to get optimum solutions. Many swarm intelligent algorithms have been designed so far which are having many limitations as some get trapped in local optima, a few are having low convergence speed, some are having poor global search facilities, etc. Still, there is a requirement of designing a new algorithm or modification of existing algorithms to overcome the limitations of the existing techniques. A new Hybrid Cat Swarm Optimization algorithm named H-CSO was designed inspired by the HEFT algorithm and the initialization problem of the Cat Swarm Optimization was overcome. Still, that algorithm has a limitation of getting stuck in local minima. To overcome this algorithm a part of the Crow Search Algorithm has been integrated into H-CSO and described in this paper. After simulation, it was found that the new hybrid algorithm named HC-CSO outperforms CSO and H-CSO.

Independent Task Scheduling in Cloud Computing using Meta-Heuristic HC-CSO Algorithm

Cloud computing is a vital paradigm of emerging technologies. It provides hardware, software, and development platforms to end-users as per their demand. Task scheduling is an exciting job in the cloud computing environment. Tasks can be divided into two categories dependent and independent. Independent tasks are not connected to any type of parent-child concept. Various meta-heuristic algorithms have come into force to schedule the independent tasks. In this, paper a hybrid HC-CSO algorithm has been simulated using independent tasks. This hybrid algorithm has been designed by using the HEFT algorithm, Self-Motivated Inertia Weight factor, and standard Cat Swarm Optimization algorithm. The Crow Search algorithm has been applied to overcome the problem of premature convergence and to avoid the H-CSO algorithm getting stuck in the local fragment. The simulation was carried out using 500-1300 random lengths independent tasks and it was found that the H-CSO algorithm has beaten PSO, ACO, and CSO algorithms whereas the hybrid algorithm HC-CSO is working fine despite Cat Swarm Optimization, Particle Swarm Optimization, and H-CSO algorithm in the name of processing cost and makespan. For all scenarios, the HC-CSO algorithm is found overall 4.15% and 7.18% efficient than the H-CSO and standard CSO respectively in comparison to the makespan and in case of computation cost minimization, 9.60% and 14.59% than the H-CSO and the CSO, respectively.

SPO: A Secure and Performance-aware Optimization for MapReduce Scheduling

MapReduce is a common framework that effectively processes multi-petabyte data in a distributed manner. Therefore, MapReduce is widely used in heterogeneous environments, such as cloud, to provide performance adequate for system needs. Despite the MapReduce benefits, tweaking the system configuration to achieve the maximum performance is still challenging and needs deep expertise. Besides, some new MapReduce security issues, which has not been well-addressed yet, are recently raised. In this paper, we present a performance-aware and secure framework, named SPO, to minimize the makespan of the tasks while considering task security constraints. Inspired by the HEFT algorithm, first, we introduce SPO, which proposes a two-stage static scheduler in Map and Reduce phases, respectively, to minimize makespan while considering network traffic. Plus, SPO∗ introduces a mathematical optimization model of the proposed scheduler aiming to estimate the system performance while considering security constraints with an error of less than 2%. The experimental results demonstrate that SPO outperforms Hadoop-stock in terms of makespan and network traffic by 29% and 31%, respectively, for the tasks running in heterogeneous environments.

GRP-HEFT: A Budget-Constrained Resource Provisioning Scheme for Workflow Scheduling in IaaS Clouds

In Infrastructure as a Service (IaaS) Clouds, users are charged to utilize cloud services according to a pay-per-use model. If users intend to run their workflow applications on cloud resources within a specific budget, they have to adjust their demands for cloud resources with respect to this budget. Although several scheduling approaches have introduced solutions to optimize the makespan of workflows on a set of heterogeneous IaaS cloud resources within a certain budget, the hourly-based cost model of some well-known cloud providers (e.g., Amazon EC2 Cloud) can easily lead to a higher makespan and some schedulers may not find any feasible solution. In this article, we propose a novel resource provisioning mechanism and a workflow scheduling algorithm, named Greedy Resource Provisioning and modified HEFT (GRP-HEFT), for minimizing the makespan of a given workflow subject to a budget constraint for the hourly-based cost model of modern IaaS clouds. As a resource provisioning mechanism, we propose a greedy algorithm which lists the instance types according to their efficiency rate. For our scheduler, we modified the HEFT algorithm to consider a budget limit. GRP-HEFT is compared against state-of-the-art workflow scheduling techniques, including MOACS (Multi-Objective Ant Colony System), PSO (Particle Swarm Optimization), and GA (Genetic Algorithm). The experimental results demonstrate that GRP-HEFT outperforms GA, PSO, and MOACS for several well-known scientific workflow applications for different problem sizes on average by 13.64, 19.77, and 11.69 percent, respectively. Also in terms of time complexity, GRP-HEFT outperforms GA, PSO and MOACS.

Parental Prioritization-Based Task Scheduling in Heterogeneous Systems

Efficient task scheduling is important for achieving high performance in heterogeneous distributed computing systems. The main focus of this research is to build a task scheduling algorithm for a heterogeneous environment. We proposed an algorithm named parental prioritization earliest finish time. It has two phases, tasks prioritization phase and processor assigning phase. In the tasks prioritization phase, tasks will schedule in parental priority queue (PPQ) on the basis of downward rank and parental priority. Task prioritization is based on the directed acyclic graph. It can schedule the task of successor row before the current row if it has less communication cost. In the processor assigning phase, the processor will allocate to the scheduled tasks obtained from PPQ keeping the computation cost to a minimum. This proposed algorithm is compared with HEFT and CPOP algorithms through graphs generated from random task graph generator and a set of tasks. The experimental results show that our proposed scheduling algorithm performs significantly better than other algorithms in terms of both cost and makespan of schedules.

A scheduling algorithm for applications in a cloud computing system with communication changes

AbstractThis paper proposes a scheduling algorithm to solve the problem of task scheduling in a cloud computing system with time‐varying communication conditions. This algorithm converts the scheduling problem with communication changes into a directed acyclic graph (DAG) scheduling problem for existing fuzzy communication task nodes, that is, the scheduling problem for a communication‐change DAG (CC‐DAG). The CC‐DAG contains both computation task nodes and communication task nodes. First, this paper proposes a weighted time‐series network bandwidth model to solve the indefinite processing time (cost) problem for a fuzzy communication task node. This model can accurately predict the processing time of a fuzzy communication task node. Second, to address the scheduling order problem for the computation task nodes, a dynamic pre‐scheduling search strategy (DPSS) is proposed. This strategy computes the essential paths for the pre‐scheduling of the computation task nodes based on the actual computation costs (times) of the computation task nodes and the predicted processing costs (times) of the fuzzy communication task nodes during the scheduling process. The computation task node with the longest essential path is scheduled first because its completion time directly influences the completion time of the task graph. Finally, we demonstrate the proposed algorithm via simulation experiments. The experimental results show that the proposed DPSS produced remarkable performance improvement rate on the total execution time that ranges between 11.5% and 21.2%. In view of the experimental results, the proposed algorithm provides better quality scheduling solution that is suitable for scientific application task execution in the cloud computing environment than HEFT, PEFT, and CEFT algorithms.

Optimal computing resource allocation algorithm in cloud computing based on hybrid differential parallel scheduling

In order to improve the ability of resource allocation and scheduling in cloud computing, optimize resource allocation and improve the efficiency of cloud computing, an optimal computing resource allocation algorithm in cloud computing based on hybrid differential parallel scheduling is proposed. In this algorithm, the models of data structure and gird structure of computing resource allocation in cloud computing are constructed, and the sample clustering analysis method of resource information flow is used to classify the attributes of computing resources; the sliding window of computing resource allocation is divided into multiple sub-windows; characteristic quantities associated with computing resource allocation attributes are selected in neighbor samples as standard vector sets for adaptive pairing; the computing resources in cloud computing are done with singular value decomposition and the resource allocation is transformed into the least square problem; the hybrid differential parallel computing method is used for optimal solution finding of resource scheduling vector set to prevent the allocation results from falling into local optimal solution, so as to improve the global convergence of resource allocation. The simulation results show that when the method proposed in this paper is used for resource allocation in clouding computing, the clustering performance is high and the convergence control ability to computing resources with different attributes is high; the allocation speedup can reach 3.67, which is improved by 14.65 and 7.43% respectively compared with that in the traditional HEFT algorithm and HCNF algorithm; when the number of allocate nodes is 100, the overhead is only 5.6, which is reduced by 14.56 and 8.33% than that in traditional HEFT algorithm and HCNF algorithm. So it shows that the proposed method has a higher practical application performance for its shorter execution time and lower overhead.

Hybrid Task Scheduling Method for Cloud Computing by Genetic and DE Algorithms

Cloud computing is a new technology which provides online services to the consumers. In order to have a high efficiency in cloud computing, proper task scheduling is required. Since the task scheduling in cloud computing is regarded as an NP complete problem, so traditional heuristic algorithms do not have the required efficiency in this environment. Therefore, recently, the majority of the proposed task scheduling algorithms have focused on hybrid meta-heuristic methods for task scheduling. In this paper, we proposed a hybrid meta-heuristic method by using HEFT algorithm. The obtained results of the simulation and statistical analysis revealed that the proposed algorithm outperforms three other heuristic and genetic algorithms in terms of the makespan in the randomly Direct Acyclic Graphs (DAGs).

Scheduling Independent Moldable Tasks on Multi-Cores with GPUs

We present a new approach for scheduling independent tasks on multiple CPUs and multiple GPUs. The tasks are assumed to be parallelizable on CPUs using the moldable model: the final number of cores allotted to a task can be decided and set by the scheduler. More precisely, we design an algorithm aiming at minimizing the makespan—the maximum completion time of all tasks—for this scheduling problem. The proposed algorithm combines a dual approximation scheme with a fast integer linear program (ILP). It determines both the partitioning of the tasks, i.e., whether a task should be mapped to CPUs or a GPU, and the number of CPUs allotted to a moldable task if mapped to the CPUs. A worst-case analysis shows that the algorithm has an approximation ratio of $\frac{3}{2} + \epsilon$ . Since the time complexity of the ILP-based algorithm could be non-polynomial, we also present a polynomial-time algorithm with an approximation ratio of $2+\epsilon$ . We complement the theoretical analysis of our two novel algorithms with a simulation study. In these simulations, we compare our algorithms to a modified version of the classical HEFT algorithm, which we adapted to handle moldable tasks. The simulation results show that our algorithm with the $\left(\frac{3}{2} + \epsilon \right)$ -approximation ratio produces significantly shorter schedules than the modified HEFT for most of the instances. In addition, our results provide evidence that our ILP-based algorithm can solve larger problem instances in a reasonable amount of time.

A Survey of Scheduling Algorithms for Heterogeneous Systems and Comparative Study of HEFT and CPOP Algorithms

A Survey of Scheduling Algorithms for Heterogeneous Systems and Comparative Study of HEFT and CPOP Algorithms

Customer Facilitated Cost-based Scheduling (CFCSC) in Cloud

Cloud computing is a prototype which takes the new form of utility computing. The on-demand flexible service and pay-per-use schemes provided by the Cloud providers attract the customers to move towards Cloud computing environment. Both the service providers and the customers will receive the economic benefits only when the available resources are correctly scheduled. Due to commercialization, the cloud environment emphasizes the need for the development of new algorithms for better economic factors. This is done by the task and data scheduler in the workflow systems. In this paper, an algorithm CustomerFacilitated Cost-based Scheduling (CFCSC) algorithmis proposed to favor the Cloud customers with economic cost. The CFCSC algorithm outstrips the HEFT in terms of load balance, minimum cost and less complex cost function.Also, both CFCSC and HEFT algorithms have the same makespan.

Based on Improved Genetic Algorithm for Task Scheduling of Heterogeneous Multi-Core Processor

This paper puts forward an improved genetic scheduling algorithm in order to improve the execution efficiency of task scheduling of the heterogeneous multi-core processor system and give full play to its performance. The attribute values and the high value of tasks were introduced to structure the initial population, randomly selected a method with the 50% probability to sort for task of individuals of the population, thus to get high quality initial population and ensured the diversity of the population. The experimental results have shown that the performance of the improved algorithm was better than that of the traditional genetic algorithm and the HEFT algorithm. The execution time of tasks was reduced.