Job Scheduling Research Articles

Approaches Used In Adapting Metaheuristic Optimization Algorithms Developed For Continuous Problems to Discrete Problems

Many real-world problems such as determining the type and number of wind turbines, facility placement problems, job scheduling problems, are in the category of combinatorial optimization problems in terms of the type of decision variables. However, since many of the evolutionary optimization algorithms are developed for solving continuous optimization problems, they cannot be directly applied to optimization problems with discrete decision variables. Therefore, the continuous decision variable values generated by these metaheuristics need to be converted to binary values using some techniques. In other words, to apply such algorithms to discrete optimization problems, it is necessary to adapt the candidate solution vectors of the algorithms to discrete values and make changes in their working structures. In this study, firstly, adaptation methods that are frequently used in previous studies in transforming metaheuristic optimization algorithms designed for the solution of continuous optimization problems into discrete optimization algorithms are explained. Then, the popular location update strategies used in solving discrete optimization problems are explained. The presented work summarizes the process of adapting continuous optimization algorithms to solve combinatorial problems step by step.

EPTAS for load balancing problem on parallel machines with a non-renewable resource

The problem considered is the non-preemptive scheduling of independent jobs that consume a resource (which is non-renewable and replenished regularly) on parallel uniformly related machines. The input defines the speed of machines, size of jobs, the quantity of the resource required by the jobs, the replenished quantities, and replenishment dates of the resource. Every job can start processing only after the required quantity of the resource is allocated to the job. The objective function is a generalization of makespan minimization and minimization of the lp-norm of the vector of loads of the machines. We present an EPTAS for this problem. Prior to our work only a PTAS was known in this non-renewable resource settings only for the special case of our problem of makespan minimization on identical machines.

A new and general stochastic parallel machine ScheLoc problem with limited location capacity and customer credit risk

Scheduling-Location (ScheLoc) problem considering machine location and job scheduling simultaneously is a relatively new and hot topic. The existing works assume that only one machine can be placed at a location, which may not be suitable for some practical applications. Besides, the customer credit risk which largely impacts the manufacturer’s profit has not been addressed in the ScheLoc problem. Therefore, in this work, we study a new and general stochastic parallel machine ScheLoc problem with limited location capacity and customer credit risk. The problem consists of determining the machine-to-location assignment, job acceptance, job-to-machine assignment, and scheduling of accepted jobs on each machine. The objective is to maximize the worst-case probability of manufacturer’s profit being greater than or equal to a given profit (referred to as the profit likelihood). For the problem, a distributionally robust chance-constrained (DRCC) programming model is proposed. Then, we develop two model-based approaches: (1) a sample average approximation (SAA) method; (2) a model-based constructive heuristic. Numerical results of 300 instances adapted from the literature show the average profit likelihood proposed by the constructive heuristic is 9.43% higher than that provided by the SAA, while the average computation time of the constructive heuristic is only 4.24% of that needed by the SAA.

A Reinforcement Learning Approach for Scheduling Problems with Improved Generalization through Order Swapping

The scheduling of production resources (such as associating jobs to machines) plays a vital role for the manufacturing industry not only for saving energy, but also for increasing the overall efficiency. Among the different job scheduling problems, the Job Shop Scheduling Problem (JSSP) is addressed in this work. JSSP falls into the category of NP-hard Combinatorial Optimization Problem (COP), in which solving the problem through exhaustive search becomes unfeasible. Simple heuristics such as First-In, First-Out, Largest Processing Time First and metaheuristics such as taboo search are often adopted to solve the problem by truncating the search space. The viability of the methods becomes inefficient for large problem sizes as it is either far from the optimum or time consuming. In recent years, the research towards using Deep Reinforcement Learning (DRL) to solve COPs has gained interest and has shown promising results in terms of solution quality and computational efficiency. In this work, we provide an novel approach to solve the JSSP examining the objectives generalization and solution effectiveness using DRL. In particular, we employ the Proximal Policy Optimization (PPO) algorithm that adopts the policy-gradient paradigm that is found to perform well in the constrained dispatching of jobs. We incorporated a new method called Order Swapping Mechanism (OSM) in the environment to achieve better generalized learning of the problem. The performance of the presented approach is analyzed in depth by using a set of available benchmark instances and comparing our results with the work of other groups.

Role of Job Scheduling and Sequencing on Service Performance of Small-scale Fashion Design Outfits in Osogbo, Osun State, Nigeria

Role of Job Scheduling and Sequencing on Service Performance of Small-scale Fashion Design Outfits in Osogbo, Osun State, Nigeria

Correlated Combinatorial Bandits for Online Resource Allocation

We study a sequential resource allocation problem where, at each round, the decision-maker needs to allocate its limited budget among different available entities. In doing so, the decision-maker obtains the reward for each entity in that round. The goal of the decision-maker is to maximize the expected cumulative reward or equivalently minimize cumulative regret over a total of T rounds. Sequential resource allocation can be modeled as a combinatorial bandit by viewing the allocation of a budget to an entity as a base arm. In the context of resource allocation, the rewards received under different budget allocations are likely to be correlated. We propose a novel correlated combinatorial bandit framework that explicitly models such correlations. We develop a novel Correlated-UCB algorithm for online resource allocation, which yields significantly reduced regret relative to correlation-agnostic algorithms. In certain cases, our proposed algorithm even achieves bounded regret, which is an order-wise reduction in the regret relative to the correlation-agnostic approach which incurs logarithmic regret under all scenarios. We validate these performance gains through experiments on several applications such as online power allocation across wireless channels, job scheduling in multi-server systems and online access point assignment.

Performance Analysis of Modified SRPT in Multiple-Processor Multitask Scheduling

In this paper we study the multiple-processor multitask scheduling problem in both deterministic and stochastic models, where each job have several tasks and is complete only when all its tasks are finished. We consider and analyze Modified Shortest Remaining Processing Time (M-SRPT) scheduling algorithm, a simple modification of SRPT, which always schedules jobs according to SRPT whenever possible, while satisfying the non-preemptive constraint. The M-SRPT algorithm is proved to achieve a competitive ratio of θ(log α + β) for minimizing response time, where α denotes the ratio between maximum job workload and minimum job workload, β represents the ratio between maximum non-preemptive task workload and minimum job workload. In addition, the competitive ratio achieved is shown to be optimal (up to a constant factor), when there are constant number of machines. We further consider the problem under Poisson arrival and general workload distribution (i. e., M/GI/N system), and show that M-SRPT achieves asymptotic optimal mean response time when the traffic intensity ρ approaches 1, if job size distribution has finite support. Beyond finite job workload, the asymptotic optimality of M-SRPT also holds for infinite job size distributions with certain probabilistic assumptions, for example, M/M/N system with finite task workload.

Learning to Dynamically Select Cost Optimal Schedulers in Cloud Computing Environments

The operational cost of a cloud computing platform is one of the most significant Quality of Service (QoS) criteria for schedulers, crucial to keep up with the growing computational demands. Several data-driven deep neural network (DNN)-based schedulers have been proposed in recent years that outperform alternative approaches by providing scalable and effective resource management for dynamic workloads. However, state-of-the-art schedulers rely on advanced DNNs with high computational requirements, implying high scheduling costs. In non-stationary contexts, the most sophisticated schedulers may not always be required, and it may be sufficient to rely on low-cost schedulers to temporarily save operational costs. In this work, we propose MetaNet, a surrogate model that predicts the operational costs and scheduling overheads of a large number of DNNbased schedulers and chooses one on-the-fly to jointly optimize job scheduling and execution costs. This facilitates improvements in execution costs, energy usage and service level agreement violations of up to 11%, 43% and 13% compared to the state-of-the-art methods.

Digital-twin-based job shop multi-objective scheduling model and strategy

ABSTRACT Job scheduling is an important part of production management and an effective method for increasing productivity, improving enterprise competitiveness, and reducing production energy consumption. However, in traditional scheduling, some problems, such as low information transparency, response lag, poor accuracy, and poor optimisation, affect the scheduling performance. Therefore, this study presents a digital-twin-based (DT-based) job shop scheduling strategy to address these problems. First, energy consumption is introduced, and a job shop scheduling model that minimises the completion time, tardiness, and energy consumption is established. Then, a scheduling decision framework based on cloud-edge computing is proposed, and the DT-based job shop system composition and operating mechanism are described. In addition, a DT-based shop scheduling strategy, comprising an overall scheduling mechanism, an accurate processing time determination method that uses a time and space compression ratio simulation, and a data comparison and abnormality determination method, is designed. Moreover, the non-dominated sorting genetic algorithm (NSGA-II) is improved by incorporating a multi-mode crossover and random variation method, along with a variable proportion elite retention strategy. Finally, the effectiveness of the improved algorithm and proposed strategy is verified using a standard dataset and practical processing problem.

On scheduling of step-improving jobs to minimize the total weighted completion time

Kim et al. (2022) studied single-machine scheduling of step-improving jobs with a common discount factor to minimize the total weighted completion time. This study proposes a pseudo-polynomial algorithm to solve the problem. We then consider the general problem in which the discount factors are job dependent. It is proven that there is an optimal solution in which the early normal jobs are sequenced in the WSPT order and the late discounted jobs are sequenced in the WSPT order. Based on two WSPT lists, a dynamic programming solution algorithm is proposed. The run time is pseudo-polynomial when the distance of any job is bounded by a constant, where the distance of a job is the absolute difference of the two positions of this job in two WSPT-ordered job lists.

Orchestration of materials science workflows for heterogeneous resources at large scale

In the era of big data, materials science workflows need to handle large-scale data distribution, storage, and computation. Any of these areas can become a performance bottleneck. We present a framework for analyzing internal material structures (e.g., cracks) to mitigate these bottlenecks. We demonstrate the effectiveness of our framework for a workflow performing synchrotron X-ray computed tomography reconstruction and segmentation of a silica-based structure. Our framework provides a cloud-based, cutting-edge solution to challenges such as growing intermediate and output data and heavy resource demands during image reconstruction and segmentation. Specifically, our framework efficiently manages data storage, scaling up compute resources on the cloud. The multi-layer software structure of our framework includes three layers. A top layer uses Jupyter notebooks and serves as the user interface. A middle layer uses Ansible for resource deployment and managing the execution environment. A low layer is dedicated to resource management and provides resource management and job scheduling on heterogeneous nodes (i.e., GPU and CPU). At the core of this layer, Kubernetes supports resource management, and Dask enables large-scale job scheduling for heterogeneous resources. The broader impact of our work is four-fold: through our framework, we hide the complexity of the cloud’s software stack to the user who otherwise is required to have expertise in cloud technologies; we manage job scheduling efficiently and in a scalable manner; we enable resource elasticity and workflow orchestration at a large scale; and we facilitate moving the study of nonporous structures, which has wide applications in engineering and scientific fields, to the cloud. While we demonstrate the capability of our framework for a specific materials science application, it can be adapted for other applications and domains because of its modular, multi-layer architecture.

An Advanced Job Scheduling Algorithmic Architecture to Reduce Energy Consumption and CO2 Emissions in Multi-Cloud

Cloud Computing is one of the emerging fields in the modern-day world. Due to the increased volume of job requests, job schedulers have received updates one at a time. The evolution of machine learning in the context of cloud schedules has had a significant impact on cost reduction in terms of energy consumption and makespan. The research article presents a two-phase process for the scheduling architecture of cloud computing where PMs are the main working unit and users are supplied to the PMs based on the work abilities of the PM in terms of resources. A minimum cost is desired in the preliminary phase of the allocation of the user to the PM. A clustered approach utilizing k-means and Q-learning was imposed to migrate the users from one PM to another PM based on Quality of Service (QoS) parameters. The proposed work has also incorporated CO2 emissions as a major evaluation parameter other than energy consumption. To support resource sharing, the deployment model is a multi-cloud model. The proposed work is evaluated against other recently proposed state of the art techniques on the basis of QoS parameters and the proposed work proved to be better in terms of efficiency at the end of the draft.

Preemptive scheduling of parallel jobs of two sizes with controllable processing times

AbstractIn parallel machine scheduling, a size of a job is defined as the number of machines that are simultaneously required for its processing. This paper considers a scheduling problem in which the set of jobs consists of jobs of two sizes: the conventional jobs (each job requires a single machine for its processing) and parallel jobs (each job to be simultaneously processed on more than one machine). The processing times are controllable, and they have to be chosen from given intervals in order to guarantee the existence of a preemptive schedule in which all jobs are completed by a common deadline. The objective is to minimize the total compression cost which reflects possible reductions in processing times. Unlike problems of classical scheduling with conventional jobs, the model under consideration cannot be directly handled by submodular optimization methods. We reduce the problem to maximizing the total weighted work of all jobs parametrized with respect to total work of the parallel jobs. The solution is delivered by separately finding the breakpoints for the maximum total weighted work of the parallel jobs and for the maximum total weighted work of the conventional jobs. This results in a polynomial-time algorithm that is no slower than needed to perform sorting of jobs’ parameters.

An asymmetric traveling salesman problem based matheuristic algorithm for flowshop group scheduling problem

The flowshop group scheduling problem (FGSP) has become a hot research problem owing to its practical applications in modern industry in recent years. The FGSP can be regarded as a combination of two coupled sub-problems. One is the group scheduling sub-problem with sequence-dependent setup times. The other is the job scheduling sub-problem within each group. A mixed integer linear programming model is built for the FGSP with the makespan criterion. Based on the problem-specific knowledge, i.e., the sequence-dependent group setup times are greater than the processing time of jobs, and the number of machines is small, the group scheduling sub-problem is approximated into an asymmetric traveling salesman problem (ATSP). Then, a matheuristic algorithm (MA) is proposed by integrating a branch-and-cut algorithm and an iterated greedy (IG) algorithm, where the branch-and-cut algorithm is used to generate the optimal Hamiltonian circuit for sub-group sequences of a group sequence obtained by the IG. On 405 test instances, the proposed MA performs significantly better than several state-of-the-art algorithms in the literature.

Dual Interactive Wasserstein Generative Adversarial Network optimized with arithmetic optimization algorithm-based job scheduling in cloud-based IoT

Dual Interactive Wasserstein Generative Adversarial Network optimized with arithmetic optimization algorithm-based job scheduling in cloud-based IoT

A systematic review on security aware real-time task scheduling

Nowadays, security and scheduling are the most crucial aspects of each distributed real-time application due to the rapidly enhanced use of such applications. This study presents a systematic literature review (SLR) on security and scheduling techniques for real-time applications in homogeneous and heterogeneous environments and to identify areas that need more attention for a wide range of applicability of real-time applications. The search procedure identified 3162 articles, and 103 best relevant articles are selected based on quality assessment scores for a brief review and discussions with the area experts or academicians. This study considers articles with a quality assessment score of more than five for a brief review. Hence, the applications related to security, scheduling, load balancing, fixed priority, task, job, and packet scheduling, and deadline constraints of soft and hard real-time systems in homogeneous and heterogeneous environments as per the state-of-the-art have been discussed. This discussion provides a better understanding of the real-time applications. The survey thus, aims to present the improvements in real-time task scheduling with security constraints in different distributed computing environments. The SLR also highlights various real-time task scheduling algorithms/techniques including the role of security. This SLR also provides meaningful and valuable study to the researchers by providing brief studies of various real-time task scheduling algorithms and security needs in tabular form.

Service Cost Effective and Reliability Aware Job Scheduling Algorithm on Cloud Computing Systems

Nowadays, the cloud computing services encounter several cloud Quality of Service challenges, such as reliability, cost, and response time. The most common mechanism to improve cloud service reliability is fault-tolerant technique. However, this reliability enhancement technique inevitably results in multiple replications, which lead to high service cost. In recognition of these challenges, we build a cloud computing systems resources management architecture. Then, we analyze the cloud service execution reliability on the physical resources of a VM and used a CUDA-enabled parallel two-dimensional LSTM to predict the software faults of a cloud VM. Thirdly, we propose an effective primary/backup cloud service cost calculation approach. To overcome the cloud service response time constraint, we integrate a response time slack factor into this method. Fourthly, we formulate the cloud service reliability and cost aware job scheduling problem, which aims at minimizing the total cloud service cost and rejection rate, and improving the system reliability. Fifthly, a heuristic greedy reliability and cost aware job scheduling (RCJS) algorithm is proposed. Finally, a performance evaluation is conducted and the experimental results demonstrate that our proposed RCJS algorithm significantly outperforms optimal redundant VM placement (OPVMP), MIN-MIN algorithms in terms of average service cost and rejection rate.

Online Scheduling Algorithm for Heterogeneous Distributed Machine Learning Jobs

Distributed machine learning (ML) has played a key role in todays proliferation of AI services. A typical model of distributed ML is to partition training datasets over multiple worker nodes to update model parameters in parallel, adopting a parameter server or AllReduce architecture. A fundamental problem in a distributed ML cluster is how to explore the demand elasticity of ML jobs and schedule them with different resource configurations, such that the average job completion time is minimized. To address it, we propose an online scheduling algorithm to decide the execution time window, the number and the type of concurrent workers and parameter servers for each job upon its arrival, with a goal of minimizing the weighted average completion time. Our online algorithm consists of (i) an online scheduling framework that groups unprocessed ML training jobs into a batch iteratively, and (ii) a batch scheduling algorithm that configures each ML job to maximize the total weight of scheduled jobs in the current iteration. Our online algorithm guarantees a good parameterized competitive ratio with polynomial time complexity. Extensive evaluations using real-world data demonstrate that it outperforms state-of-the-art schedulers in todays AI cloud systems.

Semi-Online Scheduling on Two Identical Parallel Machines with Initial-Lookahead Information

This work investigates a new semi-online scheduling problem with lookahead. We focus on job scheduling on two identical parallel machines, where deterministic online algorithms only know the information of [Formula: see text] initial jobs (i.e., the initial-lookahead information), while the following jobs still arrive one-by-one in an over-list fashion. We consider makespan minimization as the objective. The study aims at revealing the value of knowing [Formula: see text] initial jobs, which are used to improve the competitive performance of those online algorithms without such initial-lookahead information. We provide the following findings: (1) For the scenario where the [Formula: see text] initial jobs are all the largest jobs with length [Formula: see text], we prove that the classical LIST algorithm is optimal with competitive ratio [Formula: see text]; (2) For the scenario where the total length of these [Formula: see text] jobs is at least [Formula: see text], we show that any online algorithm has a competitive ratio at least 3/2, implying that the initial-lookahead knowledge is powerless since there exists a 3/2-competitive online algorithm without such information; (3) For the scenario where the total length of these [Formula: see text] jobs is at least [Formula: see text] ([Formula: see text]), we propose an online algorithm, named as LPT-LIST, with competitive ratio of [Formula: see text], implying that the initial-lookahead information indeed helps to improve the competitiveness of those online algorithms lacking such information.

Scheduling games with rank-based utilities

Job scheduling on parallel machines is a well-studied singleton congestion game. We consider a variant of this game, arising in environments with strong competition. In a scheduling game with rank-based utilities (SRBG) the players are partitioned into competition sets, and the goal of every player is to perform well relative to its competitors.We show that SRBGs are significantly different from classical job-scheduling games, and that competition may lead to poor outcomes. An SRBG need not have a pure Nash equilibrium, and best-response dynamics may not converge to a NE even if one exists. We identify several classes of games for which a NE exists and can be computed efficiently, present tight bounds on the equilibrium inefficiencies, and suggest ways to modify SRBGs in order to guarantee NE existence.Our analysis provides insights for several other congestion and cost-sharing games that have a natural ‘rank-based’ variant.