Multi-objective list scheduling of workflow applications in distributed computing infrastructures

Executing large-scale applications in distributed computing infrastructures (DCI), for example modern Cloud environments, involves optimization of several conflicting objectives such as makespan, reliability, energy, or economic cost. Despite this trend, scheduling in heterogeneous DCIs has been traditionally approached as a single or bi-criteria optimization problem. In this paper, we propose a generic multi-objective optimization framework supported by a list scheduling heuristic for scientific workflows in heterogeneous DCIs. The algorithm approximates the optimal solution by considering user-specified constraints on objectives in a dual strategy: maximizing the distance to the user’s constraints for dominant solutions and minimizing it otherwise. We instantiate the framework and algorithm for a four-objective case study comprising makespan, economic cost, energy consumption, and reliability as optimization goals. We implemented our method as part of the ASKALON environment (Fahringer et al., 2007) for Grid and Cloud computing and demonstrate through extensive real and synthetic simulation experiments that our algorithm outperforms related bi-criteria heuristics while meeting the user constraints most of the time.     

A multi-objective co-evolutionary algorithm for energy-efficient scheduling on a green data center

Nowadays, the environment protection and the energy crisis prompt more computing centers and data centers to use the green renewable energy in their power supply. To improve the efficiency of the renewable energy utilization and the task implementation, the computational tasks of data center should match the renewable energy supply. This paper considers a multi-objective energy-efficient task scheduling problem on a green data center partially powered by the renewable energy, where the computing nodes of the data center are DVFS-enabled. An enhanced multi-objective co-evolutionary algorithm, called OL-PICEA-g, is proposed for solving the problem, where the PICEA-g algorithm with the generalized opposition based learning is applied to search the suitable computing node, supply voltage and clock frequency for the task computation, and the smart time scheduling strategy is employed to determine the start and finish time of the task on the chosen node. In the experiments, the proposed OL-PICEA-g algorithm is compared with the PICEA-g algorithm, the smart time scheduling strategy is compared with two other scheduling strategies, i.e., Green-Oriented Scheduling Strategy and Time-Oriented Scheduling Strategy, different parameters are also tested on the randomly generated instances. Experimental results confirm the superiority and effectiveness of the proposed algorithm.     

Multi-objective scheduling of dynamic job shop using variable neighborhood search

Dynamic job shop scheduling that considers random job arrivals and machine breakdowns is studied in this paper. Considering an event driven policy rescheduling, is triggered in response to dynamic events by variable neighborhood search (VNS). A trained artificial neural network (ANN) updates parameters of VNS at any rescheduling point. Also, a multi-objective performance measure is applied as objective function that consists of makespan and tardiness. The proposed method is compared with some common dispatching rules that have widely used in the literature for dynamic job shop scheduling problem. Results illustrate the high effectiveness and efficiency of the proposed method in a variety of shop floor conditions.     

Modelling and heuristics of FMS scheduling with multiple objectives

The performance of a scheduling system, in practice, is not evaluated to satisfy a single objective, but to obtain a trade-off schedule regarding multiple objectives. Therefore, in this research, we make use of one of the multiple objective decision-making methods, a global criterion approach, to develop a multi-objective model for solving FMS scheduling problems with consideration of three performance measures, namely minimum mean job flow time, mean job tardiness, and minimum mean machine idle time, simultaneously. In addition, hybrid heuristics, which are a combination of two common local search methods, simulated annealing and tabu search, are also proposed for solving the addressed FMS scheduling problems. The feasibility and adaptability of the proposed heuristics are investigated through experimental results.     

Multi-objective optimization model for multi-project scheduling on critical chain

In this paper, a multi-project scheduling in critical chain problem is addressed. This problem considers the influence of uncertainty factors and different objectives to achieve completion rate on time of the whole projects. This paper introduces a multi-objective optimization model for multi-project scheduling on critical chain, which takes into consideration multi-objective, such as overall duration, financing costs and whole robustness. The proposed model can be used to generate alternative schedules based on the relative magnitude and importance of different objectives. To respond to this need, a cloud genetic algorithm is proposed. This algorithm using randomness and stability of Normal Cloud Model, cloud genetic algorithm was designed to generate priority of multi-project scheduling activities and obtain plan of multi-project scheduling on critical chain. The performance comparison shows that the cloud genetic algorithm significantly outperforms the previous multi-objective algorithm.     

多目标最优化云工作流调度进化遗传算法

为了实现云环境中科学工作流调度的执行跨度和执行代价的同步优化,提出了一种多目标最优化进化遗传调度算法MOEGA。该算法以进化遗传为基础,定义了任务与虚拟机映射、虚拟机与主机部署间的编码机制,设计了满足多目标优化的适应度函数。同时,为了满足种群的多样性,在调度方案中引入了交叉与变异操作,并使用启发式方法进行种群初始化。通过4种现实科学工作流的仿真实验,将其与同类型算法进行了性能比较。结果表明,MOEGA算法不仅可以满足工作流截止时间约束,而且在降低任务执行跨度与执行代价的综合性能方面也优于其他算法。     

An intelligent water drops-based workflow scheduling for IaaS cloud

Cloud computing is an emerging technology in a distributed environment with a collection of large-scale heterogeneous systems. One of the challenging issues in the cloud data center is to select the minimum number of virtual machine (VM) instances to execute the tasks of a workflow within a time limit. The objectives of such a strategy are to minimize the total execution time of a workflow and improve resource utilization. However, the existing algorithms do not guarantee to achieve high resource utilization although they have abilities to achieve high execution efficiency. The higher resource utilization depends on the reusability of VM instances. In this work, we propose a new intelligent water drops based workflow scheduling algorithm for Infrastructure-as-a-Service (IaaS) cloud. The objectives of the proposed algorithm are to achieve higher resource utilization and minimize the makespan within the given deadline and budget constraints. The first contribution of the algorithm is to find multiple partial critical paths (PCPs) of a workflow which helps in finding suitable VM instances. The second contribution is a scheduling strategy for PCP-VM assignment for assigning the VM instances. The proposed algorithm is evaluated through various simulation runs using synthetic datasets and various performance metrics. Through comparison, we show the superior performance of the proposed algorithm over the existing ones.     

Comparison of selected algorithms for scheduling workflow applications with dynamically changing service availability

This paper compares the quality and execution times of several algorithms for scheduling service based workflow applications with changeable service availability and parameters. A workflow is defined as an acyclic directed graph with nodes corresponding to tasks and edges to dependencies between tasks. For each task, one out of several available services needs to be chosen and scheduled to minimize the workflow execution time and keep the cost of service within the budget. During the execution ofa workflow, some services may become unavailable, new ones may appear, and costs and execution times may change with a certain probability. Rescheduling is needed to obtain a better schedule. A solution is proposed on how integer linear pro- gramming can be used to solve this problem to obtain optimal solutions for smaller problems or suboptimal solutions for larger ones. It is compared side-by-side with GAIN, divide-and-conquer, and genetic algorithms for various probabilities of service unavailability or change in service parameters. The algorithms are implemented and subsequently tested in a real BeesyCluster environment.     

Multi-objective scheduling of many tasks in cloud platforms

The scheduling of a many-task workflow in a distributed computing platform is a well known NP-hard problem. The problem is even more complex and challenging when the virtualized clusters are used to execute a large number of tasks in a cloud computing platform. The difficulty lies in satisfying multiple objectives that may be of conflicting nature. For instance, it is difficult to minimize the makespan of many tasks, while reducing the resource cost and preserving the fault tolerance and/or the quality of service (QoS) at the same time. These conflicting requirements and goals are difficult to optimize due to the unknown runtime conditions, such as the availability of the resources and random workload distributions. Instead of taking a very long time to generate an optimal schedule, we propose a new method to generate suboptimal or sufficiently good schedules for smooth multitask workflows on cloud platforms.Our new multi-objective scheduling (MOS) scheme is specially tailored for clouds and based on the ordinal optimization (OO) method that was originally developed by the automation community for the design optimization of very complex dynamic systems. We extend the OO scheme to meet the special demands from cloud platforms that apply to virtual clusters of servers from multiple data centers. We prove the suboptimality through mathematical analysis. The major advantage of our MOS method lies in the significantly reduced scheduling overhead time and yet a close to optimal performance. Extensive experiments were carried out on virtual clusters with 16 to 128 virtual machines. The multitasking workflow is obtained from a real scientific LIGO workload for earth gravitational wave analysis. The experimental results show that our proposed algorithm rapidly and effectively generates a small set of semi-optimal scheduling solutions. On a 128-node virtual cluster, the method results in a thousand times of reduction in the search time for semi-optimal workflow schedules compared with the use of the Monte Carlo and the Blind Pick methods for the same purpose.     

Multi-objective method for divisible load scheduling in multi-level tree network

There is extensive literature concerning the divisible load theory. Based on the divisible load theory (DLT) the load can be divided into some arbitrary independent parts, in which each part can be processed independently by a processor. The divisible load theory has also been examined on the processors that cheat the algorithm, i.e., the processors do not report their true computation rates. According to the literature, if the processors do not report their true computation rates, the divisible load scheduling model fails to achieve its optimal performance. This paper focuses on the divisible load scheduling, where the processors cheat the algorithm. In this paper, a multi-objective method for divisible load scheduling is proposed. The goal is to improve the performance of the divisible load scheduling when the processors cheat the algorithm. The proposed method has been examined on several function approximation problems. The experimental results indicate the proposed method has approximately 66% decrease in finish time in the best case.     

Multi-objective metaheuristic algorithms for the resource-constrained project scheduling problem with precedence relations

This study addresses the resource-constrained project scheduling problem with precedence relations, and aims at minimizing two criteria: the makespan and the total weighted start time of the activities. To solve the problem, five multi-objective metaheuristic algorithms are analyzed, based on Multi-objective GRASP (MOG), Multi-objective Variable Neighborhood Search (MOVNS) and Pareto Iterated Local Search (PILS) methods. The proposed algorithms use strategies based on the concept of Pareto Dominance to search for solutions and determine the set of non-dominated solutions. The solutions obtained by the algorithms, from a set of instances adapted from the literature, are compared using four multi-objective performance measures: distance metrics, hypervolume indicator, epsilon metric and error ratio. The computational tests have indicated an algorithm based on MOVNS as the most efficient one, compared to the distance metrics; also, a combined feature of MOG and MOVNS appears to be superior compared to the hypervolume and epsilon metrics and one based on PILS compared to the error ratio. Statistical experiments have shown a significant difference between some proposed algorithms compared to the distance metrics, epsilon metric and error ratio. However, significant difference between the proposed algorithms with respect to hypervolume indicator was not observed.     

A robust scheduling method based on a multi-objective immune algorithm

A robust scheduling method is proposed to solve uncertain scheduling problems. An uncertain scheduling problem is modeled by a set of workflow models, and then a scheduling scheme (solution) of the problem can be evaluated by workflow simulations executed with the workflow models in the set. A multi-objective immune algorithm is presented to find Pareto optimal robust scheduling schemes that have good performance for each model in the set. The two optimization objectives for scheduling schemes are the indices of the optimality and robustness of the scheduling results. An antibody represents a resource allocation scheme, and the methods of antibody coding and decoding are designed to deal with resource conflicts during workflow simulations. Experimental tests show that the proposed method can generate a robust scheduling scheme that is insensitive to uncertain scheduling environments.     

自定义服务质量模型的网格工作流调度算法

随着网格服务应用的发展,在网格工作流中,复杂的任务可以由多个独立的服务,通过工作流引擎等方式组合成新服务后完成。在组合服务的过程中,由不同服务提供商提供的候选服务,具有不同的服务质量参数,在网格工作流调度中,需要满足用户定义的服务质量约束。提出了方便用户定义的服务质量模型,并且在该模型的基础上,改进了网格工作流调度算法,通过实验分析证明改进后的算法优于传统的调度算法。     

Multi-objective evolutionary simulation-optimisation of a real-world manufacturing problem

Many real-world manufacturing problems are too complex to be modelled analytically. For these problems, simulation can be a powerful tool for system analysis and optimisation. While traditional optimisation methods have been unable to cope with the complexities of many problems approached by simulation, evolutionary algorithms have proven to be highly useful. This paper describes how simulation and evolutionary algorithms have been combined to improve a manufacturing cell at Volvo Aero in Sweden. This cell produces high-technology engine components for civilian and military airplanes, and also for space rockets. Results from the study show that by using simulation and evolutionary algorithms, it is possible to increase the overall utilisation of the cell and at the same time decrease the number of overdue components.     

Adaptive energy-efficient scheduling for real-time tasks on DVS-enabled heterogeneous clusters

Developing energy-efficient clusters not only can reduce power electricity cost but also can improve system reliability. Existing scheduling strategies developed for energy-efficient clusters conserve energy at the cost of performance. The performance problem becomes especially apparent when cluster computing systems are heavily loaded. To address this issue, we propose in this paper a novel scheduling strategy–adaptive energy-efficient scheduling or AEES–for aperiodic and independent real-time tasks on heterogeneous clusters with dynamic voltage scaling. The AEES scheme aims to adaptively adjust voltages according to the workload conditions of a cluster, thereby making the best trade-offs between energy conservation and schedulability. When the cluster is heavily loaded, AEES considers voltage levels of both new tasks and running tasks to meet tasks’ deadlines. Under light load, AEES aggressively reduces the voltage levels to conserve energy while maintaining higher guarantee ratios. We conducted extensive experiments to compare AEES with an existing algorithm–MEG, as well as two baseline algorithms–MELV, MEHV. Experimental results show that AEES significantly improves the scheduling quality of MELV, MEHV and MEG.     

Reliable workflow scheduling with less resource redundancy

We examine the problem of reliable workflow scheduling with less resource redundancy. As scheduling workflow applications in heterogeneous systems, either for optimizing the reliability or for minimizing the makespan, are NP-Complete problems, we alternatively find schedules for meeting specific reliability and deadline requirements. First, we analyze the reliability of a given schedule using two important definitions: Accumulated Processor Reliability (APR) and Accumulated Communication Reliability (ACR). Second, inspired by the reliability analysis, we present three scheduling algorithms: RR algorithm schedules least Resources to meet the Reliability requirement; DRR algorithm extends RR by further considering the Deadline requirement; and dynamic algorithm schedules tasks dynamically: It avoids the “Chain effect” caused by uncertainties on the task execution time estimates, and relieves the impact from the inaccuracy on failure estimation. Finally, the empirical evaluation shows that our algorithms can save a significant amount of computation and communication resources when performing a similar reliability compared to Fault-Tolerant-Scheduling-Algorithm (FTSA) algorithm.     

Multi-objective adaptive large neighborhood search for distributed reentrant permutation flow shop scheduling

Factory management plays an important role in improving the productivity and quality of service in the production process. In particular, the distributed permutation flow shop scheduling problem with multiple factories is considered a priority factor in the factory automation. This study proposes a novel model of the developed distributed scheduling by supplementing the reentrant characteristic into the model of distributed reentrant permutation flow shop (DRPFS) scheduling. This problem is described as a given set of jobs with a number of reentrant layers is processed in the factories, which compromises a set of machines, with the same properties. The aim of the study is to determine the number of factory needs to be used, jobs assignment to certain factory and sequence of job assigned to the factory in order to simultaneously satisfy three objectives of minimizing makespan, total cost and average tardiness. To do this, a novel multi-objective adaptive large neighborhood search (MOALNS) algorithm is developed for finding the near optimal solutions based on the Pareto front. Various destroy and repair operators are presented to balance between intensification and diversification of searching process. The numerical examples of computational experiments are carried out to validate the proposed model. The analytical results on the performance of proposed algorithm are checked and compared with the existing methods to validate the effectiveness and robustness of the proposed potential algorithm in handling the DRPFS problem.     

Multi-objective scheduling with fuzzy due-date

In this paper, we examine the characteristic features of multi-objective scheduling problems formulated with the concept of fuzzy due-date. By computer simulations, we show that various scheduling criteria can be expressed by modifying the shape of membership functions of fuzzy due-dates. We also show the difficulty in handling the minimum satisfaction grade as a scheduling criterion. This difficulty is caused by the fact that the minimum satisfaction grade is zero for almost all schedules. This makes many search algorithms inefficient. We suggest an idea to cope with this difficulty.     

网格工作流调度算法研究综述*

作为一个NP完全问题,通常采用启发式算法来解决网格工作流调度。首先对网格工作流调度算法进 行了分类,然后对其典型算法进行了分析和讨论,并阐述了一些典型网格工作流调度系统,最后指出了现有算法 中的一些不足之处,展望了该领域的进一步研究方向。     

Decision support tool for multi-objective job shop scheduling problems with linguistically quantified decision functions

This paper presents a new tool for multi-objective job shop scheduling problems. The tool encompasses an interactive fuzzy multi-objective genetic algorithm (GA) which considers aspiration levels set by the decision maker (DM) for all the objectives. The GA's decision (fitness) function is defined as a measure of truth of a linguistically quantified statement, imprecisely specified by the DM using linguistic quantifiers such as most, few, etc., that refer to acceptable distances between the achieved objective values and the aspiration levels. The linguistic quantifiers are modelled using fuzzy sets. The developed tool is used to analyse and solve a real-world problem defined in collaboration with a pottery company. The tool provides a valuable support in performing various what-if analyses, for example, how changes of batch sizes, aspiration levels, linguistic quantifiers and the measure of acceptable distances affect the final schedule.