Flow Scheduling Research Articles

TC-Flow: Chain Flow Scheduling for Advanced Industrial Applications in Time-Sensitive Networks

Time-sensitive networking (TSN) can help standardize deterministic Ethernet across industrial automation. The deterministic guarantee of TSN is based on network resource scheduling in the unit of flow. However, the state-of-the-art TSN single flow scheduling scheme cannot meet the coordinated scheduling requirements of multiple data flows in advanced industrial applications (e.g., control and safety applications). In this article, we propose a TSN chain flow abstraction, TC-Flow, for a coordinated multiple-flow scheduling model in industrial control and safety applications. Based on the proposed TC-Flow model, we design an offline TC-Flow scheduling algorithm using integer linear programming and an online heuristic TC-Flow scheduling algorithm to handle network dynamics. To deploy the proposed TC-Flow model and scheduling algorithms in the TSN, we design a CF-TSN network architecture that is compatible with the existing TSN single-flow scheduling scheme. Finally, we implement the proposed CF-TSN architecture and TC-Flow scheduling algorithms in real-world network environments. Experimental results show that the proposed scheduling algorithms can increase the number of schedulable flows by 26 percent compared to the state-of-the-art TSN scheduling benchmark.

Flow Scheduling in Data Center Networks with Time and Energy Constraints: A Software-Defined Network Approach

Flow scheduling in Data Center Networks (DCN) is a hot topic as cloud computing and virtualization are becoming the dominant paradigm in the increasing demand of digital services. Within the cost of the DCN, the energy demands associated with the network infrastructure represent an important portion. When flows have temporal restrictions, the scheduling with path selection to reduce the number of active switching devices is a NP-hard problem as proven in the literature. In this paper, an heuristic approach to schedule real-time flows in data-centers is proposed, optimizing the temporal requirements while reducing the energy consumption in the network infrastructure via a proper selection of the paths. The experiments show good performance of the solutions found in relation to exact solution approximations based on an integer linear programming model. The possibility of programming the network switches allows the dynamic schedule of paths of flows under the software-defined network management.

Efficient Dynamic Deployment of Simulation Tasks in Collaborative Cloud and Edge Environments

Cloud computing has been studied and used extensively in many scenarios for its nearly unlimited resources and X as a service model. To reduce the latency for accessing the remote cloud data centers, small data centers or cloudlets are deployed near end-users, which is also called edge computing. In this paper, we mainly focus on the efficient scheduling of distributed simulation tasks in collaborative cloud and edge environments. Since simulation tasks are usually tightly coupled with each other by sending many messages and the status of tasks and hosts may also change frequently, it is essentially a dynamic bin-packing problem. Unfortunately, popular methods, such as meta-heuristics, and accurate algorithms are time-consuming and cannot deal with the dynamic changes of tasks and hosts efficiently. In this paper, we present Pool, an incremental flow-based scheduler, to minimize the overall communication cost of all tasks in a reasonable time span with the consideration of migration cost of task. After formulating such a scheduling problem as a min-cost max-flow (MCMF) problem, incremental MCMF algorithms are adopted to accelerate the procedure of calculating an optimal flow and heuristic scheduling algorithm, with the awareness of task migration cost, designed to assign tasks. Simulation experiments on Alibaba cluster trace show that Pool can schedule all of the tasks efficiently and is almost 5.8 times faster than the baseline method when few tasks and hosts change in the small problem scale.

Fast and reliable transient simulation and continuous optimization of large-scale gas networks

We are concerned with the simulation and optimization of large-scale gas pipeline systems in an error-controlled environment. The gas flow dynamics is locally approximated by sufficiently accurate physical models taken from a hierarchy of decreasing complexity and varying over time. Feasible work regions of compressor stations consisting of several turbo compressors are included by semiconvex approximations of aggregated characteristic fields. A discrete adjoint approach within a first-discretize-then-optimize strategy is proposed and a sequential quadratic programming with an active set strategy is applied to solve the nonlinear constrained optimization problems resulting from a validation of nominations. The method proposed here accelerates the computation of near-term forecasts of sudden changes in the gas management and allows for an economic control of intra-day gas flow schedules in large networks. Case studies for real gas pipeline systems show the remarkable performance of the new method.

Intelligent Network Traffic Control Based on Deep Reinforcement Learning

This paper studies the problems of load balancing and flow control in data center network, and analyzes several common flow control schemes in data center intelligent network and their existing problems. On this basis, the network traffic control problem is modeled with the goal of deep reinforcement learning strategy optimization, and an intelligent network traffic control method based on deep reinforcement learning is proposed. At the same time, for the flow control order problem in deep reinforcement learning algorithm, a flow scheduling priority algorithm is proposed innovatively. According to the decision output, the corresponding flow control and control are carried out, so as to realize the load balance of the network. Finally, experiments show, the network traffic bandwidth loss rate of the proposed intelligent network traffic control method is low. Under the condition of random 60 traffic density, the average bisection bandwidth obtained by the proposed intelligent network traffic control method is 4.0mbps and the control error rate is 2.25%. The intelligent network traffic control method based on deep reinforcement learning has high practicability in the practical application process, and fully meets the research requirements.

DCI-NACC: flow scheduling and congestion control based on programmable data plane in high-performance data center networks

DCI-NACC: flow scheduling and congestion control based on programmable data plane in high-performance data center networks

A Twofold Model for VNF Embedding and Time-Sensitive Network Flow Scheduling

The revolution that Industrial Control Systems are undergoing is reshaping the traditional network management and it is introducing new challenges. After the advent of network virtualization, an enhanced level of automation has been required to cope with the safety-critical mission of industrial systems and strict requirements for end-to-end latency. In the literature, there have been attempts to automatically solve two strictly interconnected problems for the management of virtual industrial networks: the Virtual Network Function embedding and the time-sensitive flow scheduling problems. However, dealing with these problems separately can lead to unoptimized results, or in the worst case to situations where the latency requirements cannot be satisfied because of a poorly chosen function embedding. In light of these motivations, this paper proposes a formal approach to jointly solve the two problems through an Optimization Satisfiability Modulo Theories formulation. This choice also allows combining two vital features: a formal guarantee of the solution correctness and optimization targeting the minimization of the end-to-end delay for flow scheduling. The feasibility of the proposed approach has been validated by implementing a prototype framework and experimentally testing it on realistic use cases.

Dynamic Network Security Function Enforcement via Joint Flow and Function Scheduling

Network Function Virtualization (NFV) is a new networking paradigm to enable dynamic network function deployment in networks. Existing studies focused on optimized function deployment and management in NFV. Unfortunately, these studies did not well address the problem of efficient security function enforcement in networks, which is the goal of deploying network functions (NFs), i.e., for real-time security function enforcement on the traffic, since optimal function deployment does not mean efficient security function enforcement on network traffic. In particular, they incurred significant NF enforcement cost. In order to address this issue, in this paper, we propose <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\textsf {FuncE}}$ </tex-math></inline-formula> that aims to solve the efficient real-time security function enforcement problem by developing unified dynamic flow and function scheduling. We formulate the problem as an integer linear programming problem and prove that it is NP-hard. We tackle the problem by decomposing it and developing heuristics to achieve near-optimal solutions. We conduct comprehensive experiments by using real topologies to demonstrate the effectiveness of the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\textsf {FuncE}}$ </tex-math></inline-formula> design. The experimental results demonstrate that <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\textsf {FuncE}}$ </tex-math></inline-formula> achieves near-optimal network function enforcement, which incurs over 100 times less latency than the existing the optimal solver. In particular, compared to the state-of-art defenses, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\textsf {FuncE}}$ </tex-math></inline-formula> processes the same number of candidate flows using over 50% less VNFs, while ensuring the same level of function enforcement.

Risk Assessment of Main Electrical Connection in Substation With Regional Grid Safety Constraints

To ensure the safe and stable operation of the power grid, it is necessary to carry out the risk assessment of the main electrical connection in the substation. In traditional methods, the influences of the regional power grid on the main electrical connection scheme are ignored. This paper aims to propose a new method of main electrical connection risk assessment considering the regional grid safety constraints. Firstly, the traditional risk indexes are modified by considering the load shedding in the regional grid, and safety constraints indexes are built by using severity functions. Secondly, the analytic method is adopted to analyze the faults in the substation and determine the consequences, and the external equivalent of the substation is conducted. Finally, the regional grid’s power flow and optimal scheduling are calculated, and the risk assessment results are obtained. To verify the effectiveness of the proposed method, the IEEE-RTBS system and a real 220kV substation and its adjacent grid are taken for analysis. The results show that compared with the traditional method, the risk indexes obtained by the proposed method are more conservative, and the weakness of the main electrical connection under maintenance conditions can be obtained, which proves the proposed method’s feasibility and effectiveness.

A Novel Modeling and Scheduling Method for Modern Roundabouts

This paper investigates optimal traffic flow scheduling for signalized roundabouts. First, a novel model for roundabouts is introduced based on the discretization of traffic flow. Then, the signal control problem for signalized roundabouts is formulated and further converted into a Markov Decision Process (MDP)-based optimization problem. Some fundamental properties about this problem are revealed. It is anticipated that the proposed modeling and signal designing method can provide an effective technique for traffic flow scheduling issues.

A Blockchain-Based Authentication Scheme and Secure Architecture for IoT-Enabled Maritime Transportation Systems

Although modern Maritime Transportation Systems (MTS) have been extensively benefited from Internet of Things (IoT) technology, but still the risks and challenges in safety and reliability have increased substantially. The involvement of different maritime parties in the marine transportation flow scheduling and management further escalates these challenges. Thus, we need an IoT-based collaborative processing system that unifies the modular structure and integrates multiple modules involved in MTS. Moreover, the need for a shared and controlled access mechanism that cannot be manipulated or tampered by unauthorized parties is also essential requirement in MTS. Blockchain, as an emerging technology, has become a key tool in data security protection because of its non-tampering and non-forgery characteristics. Keeping in view of this aspect, in this paper, an IoT-based collaborative processing system based on blockchain is proposed for marine transportation flow scheduling and management. In addition, we propose a novel consensus mechanism based on Verifiable Random Function (VRF) and reputation voting to reduce the communication cost in blockchain consensus communication process. The proposed scheme has been validated in a simulated environment and the results illustrate that the scheme has obvious effect in resisting replay attack and camouflage attack. Furthermore, the optimized consensus mechanism improves the security by 8% and the transaction processing speed by 6% on the premise that the communication cost is basically unchanged.

Oddlab: fault-tolerant aware load-balancing framework for data center networks

Data center networks (DCNs) act as critical infrastructures for emerging technologies. In general, a DCN involves a multi-rooted tree with various shortest paths of equal length from end to end. The DCN fabric must be maintained and monitored to guarantee high availability and better QoS. Traditional traffic engineering (TE) methods frequently reroute large flows based on the shortest and least-congested paths to maintain high service availability. This procedure results in a weak link utilization with frequent packet reordering. Moreover, DCN link failures are typical problems. State-of-the-art approaches address such challenges by modifying the network components (switches or hosts) to discover and avoid broken connections. This study proposes Oddlab (Odds labels), a novel deployable TE method to guarantee the QoS of multi-rooted data center (DC) traffic in symmetric and asymmetric modes. Oddlab creatively builds a heuristic model for efficient flow scheduling and faulty link detection by exclusively using the gathered statistics from the DCN data plane, such as residual bandwidth and the number of installed elephant flows. Besides, the proposed method is implemented in an SDN-based DCN without altering the network components. Our findings indicate that Oddlab can minimize the flow completion time, maximize bisection bandwidth, improve network utilization, and recognize faulty links with sufficient accuracy to improve DC productivity.

Research on optimization of assembly line based on product scheduling and just-in-time feeding of parts

PurposeReducing production auxiliary time is the key to improve the efficiency of the existing mixed-flow assembly line. This paper proposes a method combining improved genetic algorithm (GA) and Flexsim software. It also investigates mixed-flow assembly line scheduling and just-in-time (JIT) parts feeding scheme to reduce waste in production while taking the existing hill-drop mixed-flow assembly line as an example to verify the effectiveness of the method.Design/methodology/approachIn this research, a method is presented to optimize the efficiency of the present assembly line. The multi-objective mathematical model is established based on the objective function of the minimum production cycle and part consumption balance, and the solution model is developed using multi-objective GA to obtain the mixed flow scheduling scheme of the hill-drop planter. Furthermore, modeling and simulation with Flexsim software are investigated along with the contents of line inventory, parts transportation means, daily feeding times and time points.FindingsTheoretical analysis and simulation experiments are carried out in this paper while taking an example of a hill-drop planter mixed-flow assembly line. The results indicate that the method can effectively reduce the idle and overload of the assembly line, use the transportation resources rationally and decrease the accumulation of the line inventory.Originality/valueThe method of combining improved GA and Flexsim software was used here for the first time intuitively and efficiently to study the balance of existing production lines and JIT feeding of parts. Investigating the production scheduling scheme provides a reference for the enterprise production line accompanied by the quantity allocation of transportation tools, the inventory consumption of the spare parts along the line and the utilization rate of each station to reduce the auxiliary time and apply practically.

DeepRLB: A deep reinforcement learning‐based load balancing in data center networks

SummaryData center networks (DCNs) are facing challenging control problems such as flow scheduling, congestion control, load balancing, and bandwidth allocation when dynamically handling heterogeneous mice and elephant flows. This paper focuses on the load balancing in DCNs as an online decision‐making problem with high complexity (NP‐hard problem). Traditional load balancing approaches are usually the heuristic algorithms relying on the operator's viewpoint and a short‐term knowledge of the traffic conditions and network environment. This paper proposes DeepRLB, a deep reinforcement learning (DRL)‐based load balancing approach for software‐defined networking‐based DCNs which exploits the deep deterministic policy gradient (DDPG) algorithm to adaptively learn the link‐weight values by observing the traffic flow characteristics. The actor and critic neural networks in the DDPG model have been designed based on two learning models: fully connected (DeepRLB‐Full) and convolutional (DeepRLB‐Conv). The output of the model is further used by the controller to determine the forwarding paths for traffic flows accordingly. We evaluated the performance of DeepRLB by comparing it with ECMP as well as a per‐packet load balancing algorithm in three different data center topologies under 30 heterogeneous traffic demand matrices. The obtained results showed that the DeepRLB algorithm outperforms the ECMP with respect to studied load balancing metrics considerably (DeepRLB‐Conv achieved the best performance). As a sample, DeepRLB‐Full and DeepRLB‐Conv were able to reach 19.14% and 24.46% reduction in the total number of over‐utilized and under‐utilized network links compared to ECMP, respectively.

Strategy optimization for virtual power plant complied with power to gas operation model

Virtual power plant (VPP) is an integrated class of integrated power plant composed of energy management system and small and micro distributed energy resources controlled by it. Its distributed energy sources can be distributed generator sets, distributed energy storage equipment or demand response resources distributed among many demand-side users. Its concept emphasizes more on the function and effect presented to the outside world, updating the operation concept and produce social and economic benefits, in line with the needs and direction of China's power development. Given the uncertainty in renewable energy output and demand response, the VPP market of gas networks, Power to Gas (P2G) equipment, flexible loads and other resources are introduced. Furthermore, VPP's optimal scheduling strategy includes considering power and gas scheduling for both power and gas flows, and studies the optimization model considering the dual energy market to maximize the profits of VPP. Through example studies, the economics and scheduling strategies of the proposed VPP are analyzed. Simulation results show that the VPP proposed in this paper can effectively reduce the proportion of renewable energy emissions reduction by selling power or P2G equipment. Moreover, it is conducive to maintaining the pressure stability and safe operation of the natural gas power grid. VPP can reduce the cost of unbalanced punishment and reduce the impact of uncertainty.

Simulation of Time Delay Characteristics of Time Sensitive Networking Based on MATLAB/Simulink

In order to realize the optimal design of time sensitive networking flow scheduling strategy and meet the communication requirements of time sensitive applications to the greatest extent, it is necessary to conduct a comprehensive study on the delay characteristics by using effective simulation methods. Firstly, the working principle of time sensitive networking flow control mechanism and the influencing factors of time delay are analyzed, and the basic goal of simulation research on time delay characteristics is defined. Then, the technical characteristics of different network simulation methods are compared, and MATLAB/Simulink is selected as the simulation tool to design the time sensitive networking delay model and simulation algorithm. Finally, aiming at the main flow control mechanism application scenarios, the feasibility and correctness of the time sensitive networking delay simulation method based on MATLAB/Simulink are verified. The research results have reference value for the study of time delay characteristics and the optimization design of flow control strategy of time sensitive networking.

Effective *-flow schedule for optical circuit switching based data center networks: A comprehensive survey

To overcome the performance bottlenecks (e.g. power consumption and bandwidth) of electrical packet switching based data center networks, Optical Circuit Switching(OCS) technology has been widely studied recently, aiming at constructing the next-generation data center and carrying the ever-increasing network traffic. To achieve this goal, how to effectively handle the emerging application-level communication in nowadays data centers, i.e. how to adapt OCS to various of traffic flows, has turned to one of the crucial topics in this area. In this paper, we summarize the state-of-the-art works that focus on the flow schedule problems in OCS based data center networks and further introduce the future directions. The scope of this paper includes the schedule problems on unicast-flows, multicast-flows, and co-flows, which are denoted as *-flows in this paper, and are the major traffic patterns in the modern data center networks. The purpose of this paper is to present a broad research guideline for the *-flow scheduling of OCS, and motivate researchers to develop more innovative proposals and algorithms.

Optimization algorithm of wireless surveillance data transmission task based on edge computing

Cloud computing and edge computing have potent capabilities and value, and the combination of edge computing and wireless transmission has become a research hotspot, making cloud computing and edge computing adapt to broader application scenarios and exert more powerful functions. Due to the relatively large data volume of terminal devices in the edge computing network, how to process the upstream data stream and efficiently and accurately realize the wireless data monitoring transmission is a problem that the edge server needs to solve. This paper analyzes the interaction and cooperation relationship between levels, proposes a scheduling model and storage strategy for uplink data flow, and uses data flow scheduling and data flow hierarchical storage algorithms to optimize the edge system of data flow scheduling and storage. Combined with the attributes of the edge computing network system, the measurement indicators of the scheduling and storage of the terminal uplink data flow in the edge computing network are modeled and analyzed. We analyze the processing time delay, cost, and optimization target for optimizing the data transmission task. The simulation experiments prove that the algorithm in this paper has advantages in solving data transmission and storage problems in the network. It can reduce costs and delays, increase throughput, ensure monitoring node access delays, and maintain edge servers and disaster monitoring data centers.

Prior credit assessment of long-term SME projects with non-standard cash flows

Over the past three decades, the relative bank loan demand has changed due to the arising small and medium-sized enterprises (SMEs). Therefore, banks in their operations face the problem of processing an ever-increasing number of loan applications. The aim of this paper is to develop an auxiliary approach to assessing the prior creditworthiness of long-term SME projects with nonstandard cash flows.This study reveals how the principles of value-based management can be incorporated into the process of borrower’s creditworthiness assessment to improve the process of screening loan applications. For this, the internal rate of return was used as a criterion for loan granting decision at the initial stage of loan underwriting.An algorithm for the preliminary evaluation of loan applications is proposed and is based on the principle of maximizing the shareholder value of banks. This algorithm helps to define the credit terms taking into consideration the distribution of positive cash flows throughout the project’s expected economic life, calculate the possible real effective interest rate concerning the borrower’s nonstandard cash flow schedule, make a rough analysis on the economic efficiency of lending and state the necessary criterion to initiate the procedure of loan underwriting for the projects with nonstandard cash flow schedules. The proposed estimation algorithm stemming from the IRR-approach for the cash flow analysis can also be initially used by a borrower as a tool for credit solvency self-testing via screening of periods with corresponding cash flows that can be used for loan servicing.

MINIMASI MAKESPAN PADA PERSOALAN PENJADWALAN ORDERED FLOWSHOP MENGGUNAKAN PSO

The production scheduling problem is in the kind of flowshop with n jobs and m machines, to get the order of the schedule for allocating operations of the jobs to the available machines so as to get the minimum total time for completion of all job or commonly called makespan. This study uses an optimization technique approach with the PSO algorithm to get minimum makespan on the ordered flowhop scheduling problem. The performance of the scheduling algorithm presented is evaluated by testing on a benchmark data set of 240 variations in the combination number of jobs and machines. The minimum measure is obtained as a result of scheduling with PSO, whose process stops at a certain iteration when in the last 10 iterations there is no change in the value of a better makespan. The performance of the PSO algorithm is efficient at regular flow scheduling with the use of the most iterations of 19 iterations and the longest execution time of 28.42 seconds or less than half a minute, namely scheduling instances with the largest number of machines and jobs. In this research, only the analysis of the resulting minimal forward and the time of execution was carried out. Further research can be extended by not only measuring the minimum makespan, such as measuring total flowtime, total tardiness, and others.