Job Completion Time Research Articles

A machine learning-based simulation metamodeling method for dynamic scheduling in smart manufacturing systems

Conventional Digital Twins (DTs) in smart manufacturing rely on complex and time-intensive simulation models, hindering real-time DT-based decision-making. However, the availability of big data in Manufacturing Execution Systems (MES) enables training different Machine Learning (ML) models for fast and accurate predictions and decision assessments. Accordingly, this paper proposes an ML-Based Simulation Metamodeling Method (MLBSM) to facilitate DT-based decision-making for dynamic production scheduling in complex Stochastic Flexible Job Shop (SFJS) environments. The proposed MLBSM integrates three modules: a novel data vectorizing method (SPBM), multi-output Adaptive Boosting Regressor (ABR) models, and a new statistical risk evaluation method. SPBM converts unstructured production log data into numerical vectors for ABR training by calculating numeric penalty scores for each job based on the position of operations in the schedule queue. Each trained ABR predicts mean job completion times for various dynamic scenarios based on shift schedules. The risk evaluation method estimates the standard deviation of job completion times and calculates the delay probability scores for each job, aiding DT in promptly evaluating production schedules. Working seamlessly together, MLBSM modules present a novel way of using production log data for ML training and ultimately bypassing several computationally intensive simulation replications. In this research, a simulation model generates the synthetic MES data, focusing on the machining process at a photolithography workstation in the semiconductor manufacturing. Experiments demonstrate the MLBSM’s accuracy and efficiency, predicting high-risk jobs with over 80% recall and being at least 70 times faster than conventional simulation runs. Sensitivity analyses also confirm the MLBSM’s consistency under different workstation conditions.

A learning-driven multi-objective cooperative artificial bee colony algorithm for distributed flexible job shop scheduling problems with preventive maintenance and transportation operations

In recent years, distributed production scheduling problems receive much attention from both scholars and practicers. Nevertheless, existing research on such problems commonly ignores machine maintenance and job transportation operations. This work proposes a distributed flexible job shop scheduling problem with preventative maintenance and transportation operations in consideration of sequence-dependent setup time. First, a mixed integer programming model is formulated to minimize maximum completion time of jobs and maximum workload of factories. Second, a learning-driven multi-objective cooperative artificial bee colony algorithm is developed to deal with the model. A Q-learning-based cooperation strategy between population and obtained non-dominated individuals is devised to strengthen the exploration and exploitation performance. Furthermore, heuristic-based population initialization, crossover operations in employed bee phase and iterated local search methods in onlooker bee phase are specially developed considering the problem characteristics. Finally, the proposed method is compared against three well-known meta-heuristics from existing literature and an exact solver CPLEX by using a set of benchmark instances. The results confirm the competitive performance of the developed model and algorithm for addressing the studied problems.

ResLake : Towards Minimum Job Latency and Balanced Resource Utilization in Geo-Distributed Job Scheduling

At internet scale companies like ByteDance, data is generated and consumed at enormously high speed by many different applications. Achieving low latency on such big data jobs is an important problem. However, the naive approach of aggregating all the data required by a job to a single location is not always feasible in a geo-distributed environment. Similarly, existing approaches in geo-distributed job scheduling often try to minimize WAN usage, which may come at the cost of latency. Another crucial element to ensure low latency is resource load balancing among DCs, which enables flexibility in job scheduling and avoids resource bottlenecks. Therefore, to minimize latency, optimizing job completion time (JCT) while maintaining resource utilization balance is important. To this end, we propose ResLake , a global scheduling platform for data-intensive workloads. ResLake aims to reduce JCT of geo-distributed applications while balancing the compute (CPU/Memory) and storage (Disk) usages across DCs and efficiently using WAN interconnections. We have deployed ResLake in ByteDance's production for over 1.5 years. ResLake has scheduled billions of jobs since its deployment. We find that ResLake improves JCT of jobs by at least 20%, and can improve resource utilization balance across DCs by up to 53%.

DLRover-RM: Resource Optimization for Deep Recommendation Models Training in the Cloud

Deep learning recommendation models (DLRM) rely on large embedding tables to manage categorical sparse features. Expanding such embedding tables can significantly enhance model performance, but at the cost of increased GPU/CPU/memory usage. Meanwhile, tech companies have built extensive cloud-based services to accelerate training DLRM models at scale. In this paper, we conduct a deep investigation of the DLRM training platforms at AntGroup and reveal two critical challenges: low resource utilization due to suboptimal configurations by users and the tendency to encounter abnormalities due to an unstable cloud environment. To overcome them, we introduce DLRover, an elastic training framework for DLRMs designed to increase resource utilization and handle the instability of a cloud environment. DLRover develops a resource-performance model by considering the unique characteristics of DLRMs and a three-stage heuristic strategy to automatically allocate and dynamically adjust resources for DLRM training jobs for higher resource utilization. Further, DLRover develops multiple mechanisms to ensure efficient and reliable execution of DLRM training jobs. Our extensive evaluation shows that DLRover reduces job completion times by 31%, increases the job completion rate by 6%, enhances CPU usage by 15%, and improves memory utilization by 20%, compared to state-of-the-art resource scheduling frameworks. DLRover has been widely deployed at AntGroup and processes thousands of DLRM training jobs on a daily basis. DLRover is open-sourced and has been adopted by 10+ companies.

Memetic discrete differential evolution with domain knowledge for blocking scheduling distributed flow shop with lot-streaming constraints

This article addresses the blocking distributed flow shop scheduling problem with lot-streaming (BDFSPL) with the objective of minimizing the maximum completion time, or makespan. In many realistic conditions, a job is divided into multiple continuous processed sublots. This pattern achieves operation overlap on consecutive machines and decreases the completion time of the whole job. Moreover, there are no buffers between machines, potentially causing finished sublots to be blocked on the current machine until the downstream machine is available. This is the first study to consider distributed production, blocking constraints, and lot-streaming together. To address this problem, we formulate a mathematical model for BDFSPL and propose a memetic discrete differential evolution (MDDE) algorithm. First, the MDDE algorithm partitions the entire population into three subpopulations based on solution quality, and each subpopulation evolves independently through adaptive scale and parameter adjustment. Second, four discrete mathematical operators are proposed to generate new individuals, and based on these operators, discrete mutation, crossover, and selection operations of MDDE are designed. Additionally, a crucial path-based local search strategy, informed by the domain knowledge of the best individual, is embedded into MDDE to enhance the exploitation capacity. To validate MDDE’s effectiveness, comprehensive experiments with several related algorithms on extensive testing instances are carried out, and the parameters of the MDDE algorithm are calibrated by employing a design of experiments. The comparison results demonstrate the effectiveness and efficiency of the proposed MDDE algorithm for the BDFSPL.

Design of an Iterative Method for MapReduce Scheduling Using Deep Reinforcement Learning and Anomaly Detection

Due to high complexities within distributed computing environments, there's a critical need for advanced scheduling frameworks that are capable of optimizing MapReduce systems. Current approaches have static policies that limit their capability to adapt to changing system dynamics and workload variations for different cloud scenarios. To overcome these issues, this study introduces a robust MapReduce framework empowered by intelligent scheduling algorithms, tailored to enhance system efficiency and resilience levels. The framework introduces three novel scheduling models: Deep Reinforcement Learning for Dynamic Job Scheduling (DRDJS), Anomaly Detection-driven Adaptive Scheduling (ADAS), and Cluster-based Job Categorization and Scheduling (CJCS). DRDJS utilizes deep reinforcement learning to dynamically generate optimal scheduling policies based on multiple metrics that include historical job data, system metrics, and workload characteristics. This adaptive approach leads to significant reductions in job completion times, outperforming static scheduling methods by up to 30%. Next, ADAS leverages anomaly detection to prioritize critical tasks and efficiently allocate resources in response to anomalies, resulting in a notable improve scheduling speed by up to 40%. Finally, CJCS employs clustering algorithms to categorize jobs based on their resource requirements and execution profiles, enabling more accurate resource allocation and reducing average job completion times by up to 25%. By integrating these methods into a unified scheduling framework, the proposed solution addresses the variability in job characteristics and system performance, thereby enhancing the overall throughput and stability of MapReduce operations.

Scheduling maintenance activities subject to stochastic job-dependent machine deterioration

This paper considers a maintenance scheduling problem on a single machine with a fixed job sequence involving job-dependent, stochastic deterioration. If the machine breaks down, then a time-consuming emergency maintenance activity needs to be conducted. Instead, planned maintenance activities can be conducted between jobs in order to improve the machine’s state and, thus, prevent the machine from breaking down. We distinguish two types of machines differing in the flexibility regarding the point of time when the maintenance activities need to be scheduled. The goal is to minimize the total expected completion time of jobs. We formalize two optimization problems, analyze their computational complexities, develop exact solution approaches, and conduct a computational study. Here, we show that our approaches can solve large instances to optimality within short run times and we analyze the advantage gained by planning maintenance activities depending on the actual state of the machine.

ESEN: Efficient GPU sharing of Ensemble Neural Networks

Ensemble neural networks are widely applied in cloud-based inference services due to their remarkable performance, while the growing demand for low-latency services leads researchers to pay more attention to the execution efficiency of these models, especially the device utilization. It is highly desirable to fully utilize GPUs by multiplexing different inference tasks on the same GPU with advanced sharing technique, such as Multi-Process-Service (MPS). However, we find it struggling when applying MPS to Ensemble Neural Networks, which consist of multiple related sub-models. The critical challenge in this predicament revolves around the efficient allocation of resources within an ensemble, aiming to minimize job completion time.To tackle this challenge, we initially examine the interplay among individual neural networks within an ensemble, outlining a guideline for achieving the shortest job completion time. Subsequently, we establish a mathematical model to formalize the resource requirements of each neural network. We introduce a search-based allocation algorithm designed to swiftly identify optimal solutions. Finally, we introduce ESEN, comprising the search-based resource allocation algorithm and efficient model execution mechanisms within PyTorch. ESEN is augmented with customized execution mechanisms for user-friendly implementation. Experimental results demonstrate that proposed ESEN can attain an efficiency improvement up to 17.84% and a GPU utilization increase of 28.09% compared to the default strategy. With the optimization of GPU resource allocation, ESEN significantly improves the efficiency of ensemble models. It provides a low-latency and high-accuracy solution for online interactive services.

How has the COVID-19 pandemic shaped behavior in crowdsourcing? The role of online labor market training

AbstractThe outbreak of the COVID-19 pandemic and the associated lockdown measures have been a shock to market systems worldwide, affecting both the supply and demand of labor. Intensified by this pandemic-driven recession, online labor markets are in many ways at the core of the economic and policy debates about their technological innovation, which could be used as a way of economic reform and recovery. In this work, we focus on crowdsourcing, which is a specific type of online labor. We apply a unique dataset of labor data to investigate the effects of online training, a policy that was provided to requesters by the platform during the COVID-19 period. Our findings suggest that workers indirectly finance on-the-job online training by accepting lower wages during the pandemic. By utilizing a difference-in-difference research design, we also provide causal evidence that online training results in lower job completion time and the probability of being discontinued. Our findings show that both employers and employees in our online labor context reacted to the pandemic by participating in online labor procedures with different risk strategies and labor approaches. Our findings provide key insights for several groups of crowdsourcing stakeholders, including policy-makers, platform owners, hiring managers, and workers. Managerial and practical implications in relation to how online labor markets react to external shocks are discussed.

An asymptotically optimal solution for the minimization of the variation of job completion times in two-stage proportionate no-wait flow shops

We propose an asymptotically optimal solution for the two-stage no-wait proportionate flow shop scheduling problem with the objective of minimizing the total absolute deviation of job completion times. The proposed solution is within 1% of the optimum for problems with more than 20 jobs. We also show that the problem is solved optimally in constant time for the special case where the two smallest jobs have equal processing times. Our results provide answers to most of the outstanding research questions for this problem.

A Proposal for a Tokenized Intelligent System: A Prediction for an AI-Based Scheduling, Secured Using Blockchain

Intelligent systems are being proposed every day as advances in cloud systems are increasing. Mostly, the services offered by these cloud systems are dependent only on their providers, without the inclusion of services from other providers, specialized third parties, or individuals. This ‘vendor lock-in’ issue and the limitations related to offering tailored services could be resolved by allowing multiple providers or individuals to collaborate through intelligent task scheduling. To address such real-world systems’ limitations in provisioning and executing heterogeneous services, we employed Blockchain and Deep Reinforcement Learning here; the first is used for the token-based secured communication between parties, and the latter is to predict the appropriate task scheduling; hence, we guarantee the quality of not only the immediate decision but also the long-term. The empirical results show a high reward achieved, meaning that it accurately selected the candidates and adaptably assigned the tasks based on job nature and executors’ individual computing capabilities, with 95 s less than the baseline in job completion time to maintain the Quality of Service. The successful collaboration between parties in this tokenized system while securing transactions through Blockchain and predicting the right scheduling of tasks makes it a promising intelligent system for advanced use cases.

Optimal policies for minimizing total job completion times and deviations from common due dates in unrelated parallel machine scheduling

Optimal policies for minimizing total job completion times and deviations from common due dates in unrelated parallel machine scheduling

Multi-objective elitist spotted hyena resource optimized flexible job shop scheduling

The job shop scheduling problem (JSSP) has drained a lot of consideration since it is one of the most important optimization problems in the manufacturing domain. The scheduling method is crucial for optimizing the objective of minimizing makespan among thousands of jobs, but evaluating machine capacity for achieving this goal remains challenging despite the development of various population-based optimization algorithms for job shop scheduling problems. To improve the efficiency of Job shop scheduling, a novel Multi-objective Elitist Spotted Hyena Monotonic Scheduling (MESHS) technique is introduced. The proposed MESHS technique includes two major processes: machine selection and operation sequences. The number of jobs is considered for solving the scheduling problem. First, the machine selection is performed by applying the Multi-objective Elitist Spotted Hyena optimization technique. The optimization technique selects the optimal machines parallelly based on multiple objective functions such as energy consumption, CPU utilization, and job completion time. The fitness of every machine is calculated based on these multiple objective functions using Levenberg–Marquardt method. Then the Elitist strategy is applied to select the optimal machine based on fitness. After the machine selection, the rate-monotonic preemptive scheduling is modeled to provide a robust operation sequence by assigning high-priority jobs to the optimal machines. As a result, efficient job scheduling is achieved with minimum time. Finally, the experimental valuation is carried out using a benchmark OR-Library dataset with different factors such as job shop scheduling efficiency, job scheduling time, makespan, and memory consumption concerning a number of jobs.

Approximation of the Objective Function of Single-Machine Scheduling Problem

The problem of the approximation of the coefficients of the objective function of a scheduling problem for a single machine is considered. It is necessary to minimize the total weighted completion times of jobs with unknown weight coefficients when a set of problem instances with known optimal schedules is given. It is shown that the approximation problem can be reduced to finding a solution to a system of linear inequalities for weight coefficients. For the case of simultaneous job release times, a method for solving the corresponding system of inequalities has been developed. Based on it, a polynomial algorithm for finding values of weight coefficients that satisfy the given optimal schedules was constructed. The complexity of the algorithm is O(n2(N+n)) operations, where n is the number of jobs and N is the number of given instances with known optimal schedules. The accuracy of the algorithm is estimated by experimentally measuring the function ε(N,n)=1n∑j=1n∣wj−wj0∣wj0, which is an indicator of the average modulus of the relative deviation of the found values wj from the true values wj0. An analysis of the results shows a high correlation between the dependence ε(N,n) and a function of the form α(n)/N, where α(n) is a decreasing function of n.

Eventually-consistent federated scheduling for data center workloads

Data center schedulers operate at unprecedented scales today to accommodate the growing demand for computing and storage power. The challenge that schedulers face is meeting the requirements of scheduling speeds despite the scale. To do so, most scheduler architectures use parallelism. However, these architectures consist of multiple parallel scheduling entities that can only utilize partial knowledge of the data center’s state, as maintaining consistent global knowledge or state would involve considerable communication overhead. The disadvantage of scheduling without global knowledge is sub-optimal placements—tasks may be made to wait in queues even though there are resources available in zones outside the scope of the scheduling entity’s state. This leads to unnecessary queuing overheads and lower resource utilization of the data center.In this paper, we extend our previous work on Thiyyakat et al. (2023), a federated decentralized data center scheduling architecture that uses eventual consistency. The architecture utilizes both parallelism and an eventually consistent global state in each of its scheduling entities to make fast decisions in a scalable manner. In our work, we compare Megha with 3 scheduling architectures — Sparrow, Eagle, and Pigeon, using simulation. We also evaluate Megha’s prototype on a 123-node cluster and compare its performance with Pigeon’s prototype using cluster traces. The results of our experiments show that Megha consistently reduces delays in job completion time when compared to other architectures.

Flexible silicon photonic architecture for accelerating distributed deep learning

The increasing size and complexity of deep learning (DL) models have led to the wide adoption of distributed training methods in datacenters (DCs) and high-performance computing (HPC) systems. However, communication among distributed computing units (CUs) has emerged as a major bottleneck in the training process. In this study, we propose Flex-SiPAC, a flexible silicon photonic accelerated compute cluster designed to accelerate multi-tenant distributed DL training workloads. Flex-SiPAC takes a co-design approach that combines a silicon photonic hardware platform with a tailored collective algorithm, optimized to leverage the unique physical properties of the architecture. The hardware platform integrates a novel wavelength-reconfigurable transceiver design and a micro-resonator-based wavelength-reconfigurable switch, enabling the system to achieve flexible bandwidth steering in the wavelength domain. The collective algorithm is designed to support reconfigurable topologies, enabling efficient all-reduce communications that are commonly used in DL training. The feasibility of the Flex-SiPAC architecture is demonstrated through two testbed experiments. First, an optical testbed experiment demonstrates the flexible routing of wavelengths by shuffling an array of input wavelengths using a custom-designed spatial-wavelength selective switch. Second, a four-GPU testbed running two DL workloads shows a 23% improvement in job completion time compared to a similarly sized leaf-spine topology. We further evaluate Flex-SiPAC using large-scale simulations, which show that Flex-SiPAC is able to reduce the communication time by 26% to 29% compared to state-of-the-art compute clusters under representative collective operations.

Real-Time Information Access in Urban Environments: A User Interaction Study Using the Real-Time Information Test

In this study, "Real-Time Information Access in Urban Environments: A User Interaction Study Using the Real-Time Information Test," participant data revealed a diverse group with an average age of 31, a balanced gender distribution, varying education levels (40% Bachelor's, 20% Master's, 40% PhD), and an average of 6 years of experience with urban navigation. The findings of the Real-Time Information Test (RTIT) showed an average job completion time of 140 seconds and a low average error count of 1.2, demonstrating competency in interacting with real-time information systems. Furthermore, the User Satisfaction Survey found an average of 8.4 overall satisfaction ratings, 8.4 user-friendliness ratings, and 7.8 information accuracy ratings, indicating excellent user experiences. These results highlight user variety, increases in job efficiency and accuracy, and high user satisfaction, all of which contribute to a comprehensive knowledge of real-time information access in urban contexts, with implications for system advancements and urban planning.

Evaluating procedures in the NEH heuristic for the PFSP - SIST

The development and assessment of 48 heuristics for the sequence-independent setup time permutation flow shop problem (PFSP-SIST) are presented in this article. This contribution combines four tie-breaking solutions with twelve priority rules for the NEH heuristic fourth and first stage, respectively. Heuristics are evaluated on Ruiz and Allahverdi (2007) benchmark problem instances, that covers small, medium and large-size problems. The popular accelerations of Taillard were used in all tests, which were adapted to the sequence-independent setup time constraint. The aim is to reduce the longest job completion time, which is also referred to as makespan. Computation results show that using different tie-breaking strategies has a greater impact on performance than using different priority rules. The heuristics that presented the best results in relatively low computation time are those that use the FFs tie-breaking strategy procedure to the sequence-independent setup time problem.

Coalition theory based task scheduling algorithm using DLFC‐NN model

SummaryResource management and job scheduling are essential in today's cloud computing world. Due to task scheduling and users' diverse submission of large‐scale requests, co‐located VM instances negatively impacted the performance of leased VM instances. This workload further led to resource rivalry across co‐located VMs. In order to address the aforementioned problems, numerous strategies have been presented, however, they fail to take the asynchronous nature of the cloud environment into account. To address this issue, a novel “CTA using DLFC‐NN model” is proposed. This proposed approach combines the coalition theory and DLFC‐NN techniques by including IRT‐OPTICS for task size clustering, digital metrology based on ionized information (DMBII) for defect detection in virtue machines (VM), and the dynamic levy flight hamster optimization algorithm for processing time optimization of the clusters. However, the implementation of task scheduling in an online environment is limited by a number of presumptions or oversimplifications made by current scheduling systems. As a result, a unique coalition theory is applied to efficiently schedule activities. In addition, the DLFC‐NN model is used to reduce resource consumption, span time, and be highly accurate and energy‐efficient when working on both online and offline jobs. Nevertheless, while optimizing the clusters' overall execution time, earlier approaches only decreased the make‐span time for task scheduling. However, the DLFC‐NN model solves the computation problem by using a fully weighted bipartite graph and the pseudo method to determine the fitness of the least makespan time. The enhanced methodology used in this study reduces the scheduling cost and minimizes job completion times according to different task counts when compared to the existing techniques.

Approximation Algorithms for Multiprocessor Scheduling with Testing to Minimize the Total Job Completion Time

Approximation Algorithms for Multiprocessor Scheduling with Testing to Minimize the Total Job Completion Time