Integrated Resource Scheduling Research Articles

Integrating Two-Level Reinforcement Learning Process for Enhancing Task Scheduling Efficiency in a Complex Problem-Solving Environment

Task scheduling scenarios require the system designers to have complete information about the resources and their capabilities, along with the tasks and their application-specific requirements. An effective task-to-resource mapping strategy will maximize resource utilization under constraints, while minimizing the task waiting time, which will in-turn maximize the task execution efficiency. In this work, a two-level reinforcement learning algorithm for task scheduling is proposed. The algorithm utilizes a deep-intensive learning stage to generate a deployable strategy for task-to-resource mapping. This mapping is re-evaluated at specific execution breakpoints, and the strategy is re-evaluated based on the incremental learning from these breakpoints. In order to perform incremental learning, real-time parametric checking is done on the resources and the tasks; and a new strategy is devised during execution. The mean task waiting time is reduced by 20% when compared with standard algorithms like Dynamic and Integrated Resource Scheduling, Improved Differential Evolution, and Q-learning-based Improved Differential Evolution; while the resource utilization is improved by more than 15%. The algorithm is evaluated on datasets from different domains like Coronavirus disease (COVID-19) datasets of public domain, National Aeronautics and Space Administration (NASA) datasets and others. The proposed method performs consistently on all the datasets.

A dynamic approach for selecting load balancing algorithm in cloud

The collection of interconnected computers that constitutes more than one united computing resource is known as the Cloud. In recent years, the advancement of cloud computing has facilitated the rapid arrangement of interconnected data centers that are geographically dispersed, offering high-quality and dependable services. Scalable traffic management has recently been developed in cloud data centers for traffic load balancing and quality of service provisioning. However, reducing latency during multidimensional resource allocation still remains a challenge. Hence, there is a need for efficient resource scheduling to ensure load optimization in the cloud. The objective of this work is to introduce an integrated resource scheduling and load balancing algorithm for efficient cloud service provisioning. The goal of the proposed system is to select the required load balancing algorithm to enhance resource utilization. Simulations were conducted to evaluate effectiveness using the Cloudsim simulator in cloud data centers, and the results show that the proposed method achieves better performance in terms of the average success rate, resource scheduling efficiency, and response time. The dynamic nature of cloud environments requires constant adaptation in resource allocation strategies. This necessitates the development of algorithms capable of handling diverse workloads efficiently. Additionally, the increasing complexity of applications and services hosted on the cloud demands a comprehensive approach that considers not only load balancing but also the intricacies of resource utilization. Furthermore, the proposed algorithm focuses on predictive analytics to anticipate fluctuations in demand and adjust resource allocation preemptively. By incorporating machine learning techniques, the system can adapt to changing patterns, ensuring optimal performance even in unpredictable scenarios. This holistic approach addresses the evolving challenges in cloud computing, providing a robust foundation for reliable and efficient service provisioning.

Improvement in Task Scheduling Capabilities for SaaS Cloud Deployments Using Intelligent Schedulers

Task scheduling on the cloud involves processing a large set of variables from both the task side and the scheduling machine side. This processing often results in a computational model that produces efficient task to machine maps. The efficiency of such models is decided based on various parameters like computational complexity, mean waiting time for the task, effectiveness to utilize the machines, etc. In this paper, a novel Q-Dynamic and Integrated Resource Scheduling (DAIRS-Q) algorithm is proposed which combines the effectiveness of DAIRS with Q-Learning in order to reduce the task waiting time, and improve the machine utilization efficiency. The DAIRS algorithm produces an initial task to machine mapping, which is optimized with the help of a reward & penalty model using Q-Learning, and a final task-machine map is obtained. The performance of the proposed algorithm showcases a 15% reduction in task waiting time, and a 20% improvement in machine utilization when compared to DAIRS and other standard task scheduling algorithms.

Resource scheduling algorithm with load balancing for cloud service provisioning

Cloud computing uses scheduling and load balancing for virtualized file sharing in cloud infrastructure. These two have to be performed in an optimized manner in cloud computing environment to achieve optimal file sharing. Recently, Scalable traffic management has been developed in cloud data centers for traffic load balancing and quality of service provisioning. However, latency reducing during multidimensional resource allocation still remains a challenge. Hence, there necessitates efficient resource scheduling for ensuring load optimization in cloud. The objective of this work is to introduce an integrated resource scheduling and load balancing algorithm for efficient cloud service provisioning. The method constructs a Fuzzy-based Multidimensional Resource Scheduling model to obtain resource scheduling efficiency in cloud infrastructure. Increasing utilization of Virtual Machines through effective and fair load balancing is then achieved by dynamically selecting a request from a class using Multidimensional Queuing Load Optimization algorithm. A load balancing algorithm is then implemented to avoid underutilization and overutilization of resources, improving latency time for each class of request. Simulations were conducted to evaluate the effectiveness using Cloudsim simulator in cloud data centers and results shows that the proposed method achieves better performance in terms of average success rate, resource scheduling efficiency and response time. Simulation analysis shows that the method improves the resource scheduling efficiency by 7% and also reduces the response time by 35.5 % when compared to the state-of-the-art works.

Toward D2D-enhanced heterogeneous networks

In this paper, we examine upcoming 5G networks where the support of D2D communication is expected to be a key asset for operators and users alike. First, we argue the need to functionally integrate D2D and infrastructure-to-device (I2D) modes. Next, we address practical issues such as integrated resource scheduling of D2D communication within heterogeneous networks, proposing an extension of the proportional fairness algorithm that we call multi-modal proportional fairness (MMPF). We evaluate the impact of D2D in a two-tier scenario combining macroand micro-coverage, finding that although I2D retains a clear edge for general-purpose downloading, D2D is an appealing solution for localized transfers as well as for viral content.

Demonstration of joint resource scheduling in an optical network integrated computing environment [Topics in Optical Communications

Currently, optical networks have been employed to meet the ever-increasing data transfer demands of distributed scientific applications. In order to provide better performance and achieve higher resource utilization, a new networking solution and resource scheduling scheme are needed. In this article we develop an experimental network called Terabits Optical Network Integrated Computing Environment and design a joint resources scheduling system for resource management, task scheduling, and application execution. Two resource scheduling schemes, overlay scheduling and integrated scheduling, are proposed and studied. Live demonstrations and performance evaluations are conducted on the TONICE experimental network. The experimental result shows the feasibility of JRSS for real deployment. In addition, the integrated resource scheduling scheme not only reduces the execution time, but also improves the resource efficiency for distributed scientific applications.

Integrated Resource Scheduling and Bidding in the DeregulatedElectric Power Market: New Challenges

Many challenging issues arise in the newly deregulated competitive electric power markets worldwide. Instead of centralized decision-making in a monopoly environment as in the past, many parties with different goals are now involved and competing in the market. The information available to a party may be limited, regulated, and received with time delay, and decisions made by one party may influence the decision space and well-being of others. These difficulties are compounded by the underlying uncertainties inherent in the system such as the demand for electricity, fuel prices, outages of generators and transmission lines, tactics by certain market participants, etc. Consequently the market is full of uncertainty and risk. Key questions to be addressed include how to predict load and market clearing prices, how to consider other parties‘ decisions in deciding one‘s own bids, and how to manage uncertainty and risk. Since finding an optimal solution to a traditional unit commitment problem is NP-hard even without considering multiple parties and uncertainties, it may be more practical to know which decision is good with confidence rather than looking for an optimal solution. For an energy supplier, bidding decisions are coupled with generation resource scheduling or unit commitment since generator characteristics and how they will be used to meet the accepted bids in the future have to be considered before bids are submitted. For example, if starting up a thermal unit is expected, the associated startup cost should somehow be configured in the bid curves. The decisions, however, can be quite subtle since generally startup costs are not part of a bid. Bidding decisions should therefore be carefully made by considering the anticipated MCP, system demand, generation and startup costs, and competitor‘ decisions. What further complicates the issue is that some of the information is not available, or will be available but with significant delays. In paper, two promising bidding strategies for power suppliers are discussed. The ordinal optimization method seeks ’’good enough‘‘ bids with high probability, and is an effective in handling market uncertainties with much reduced computational efforts. The basic idea of this method is to use a model to describe the influence of bidding strategies on the MCP. A nominal bid curve is obtained by solving optimal power generation for a given set of the MCPs within the Lagrangian relaxation framework. Then N bids are generated by perturbing the nominal bid curve. The ordinal optimization method is applied to form a good enough bid set S, which contains some good bids with high probability, by performing rough evaluation. The best bid is then searched and selected over S by solving generation scheduling or unit commitment problems within reasonable computational time. The game theoretic method aims at bidding and self-scheduling of a utility company in New England. The problem is investigated from the viewpoint of a particular utility bidder. The uncertainties caused by bids from other bidders and the ISO (Independent System Operator) bid selection process are explicitly considered. The problem is then solved within a reduced game theoretical framework, where the ISO has a closed-form solution for a given probabilistic description of the bids, and the utility‘s problem is solved by using Lagrangian relaxation. Although the two specific methods represent significant progress made thus far, the area is wide open for creative research to make the deregulated market a true success.

Integrated resource scheduling in a distributed scheduling environment

The Space Station era presents a highly complex multimission planning and scheduling environment exercised over a highly distributed system. In order to automate the scheduling process, customers require a mechanism for communicating their scheduling requirements to NASA. An expressive scheduling notation that captures a wide range of customer requirements and scheduling options is one solution to this problem. This article describes a request language that a remotely located customer can use to specify his scheduling requirements to a NASA scheduler, thus automating the customer-scheduler interface. Additionally, this article describes a scheduler that can accept these requests, process them, generate schedules, and return schedule and resource availability information to the requester.