Dynamic Resource Optimization Research Articles

AI-Powered Predictive Analytics for Dynamic Cloud Resource Optimization: A Technical Implementation Framework

This article explores the transformative impact of AI-driven predictive analytics on cloud resource optimization, presenting a comprehensive technical implementation framework. The article shows how artificial intelligence and machine learning architectures revolutionize traditional cloud management approaches through advanced predictive capabilities and dynamic resource allocation. It examines the evolution from static threshold-based systems to sophisticated AI-driven solutions, analyzing their implementation strategies across various organizational contexts. The article delves into multiple optimization domains, including capacity provisioning, cost management, performance enhancement, and energy efficiency, while presenting real-world applications and impact analyses across different industries. Through extensive case studies and empirical evidence, the article demonstrates how organizations leverage AI-powered solutions to address complex cloud resource management challenges, achieve operational efficiencies, and maintain competitive advantages in the digital marketplace. The article also explores future developments and provides strategic recommendations for organizations implementing cloud optimization frameworks, emphasizing the importance of standardized approaches, stakeholder engagement, and sustainable practices in cloud resource management.

Dynamic Resource Optimization for Quality of Service-Driven Cost Minimization in Cognitive Radio Networks

Dynamic Resource Optimization for Quality of Service-Driven Cost Minimization in Cognitive Radio Networks

AI-Driven Process Design for Closed-Loop Manufacturing

This study investigates the potential of artificial intelligence (AI) in enhancing closed-loop manufacturing (CLM) systems, which emphasize sustainability through recycling, reuse, and optimized resource utilization. CLM faces significant challenges in minimizing waste, improving operational efficiency, and optimizing resource usage. Although AI holds promise in addressing these challenges, its application in CLM systems, especially in dynamic decision-making and resource optimization, requires further exploration. This research aims to explore how AI technologies can improve material flows, reduce waste, and enhance decision-making to achieve both sustainability and operational efficiency in manufacturing systems. To understand the impact of AI on CLM, the study conducted a comprehensive review of case studies and literature across various industries. It focused on AI techniques, such as machine learning (ML) and reinforcement learning (RL), and their roles in optimizing resource usage, refining recycling processes, and improving production schedules. Statistical methods, including regression and sensitivity analysis, were employed to evaluate the effectiveness of AI-driven models in enhancing closed-loop manufacturing performance. The review revealed promising results, highlighting AI's ability to optimize resource consumption and waste management strategies across diverse manufacturing environments. The study’s findings demonstrated significant improvements in resource efficiency, waste reduction, and cost savings across multiple industries. AI-driven models optimized material consumption, resulting in reductions of up to 12% in raw material use. Additionally, AI-enhanced recycling and waste management strategies led to a 25% increase in material recovery, while operational costs were reduced by up to 15%. AI models also displayed a strong ability to adapt to fluctuations in demand and production conditions, ensuring sustained operational efficiency even amid dynamic market changes. These outcomes highlight AI’s transformative potential in achieving sustainable manufacturing practices and advancing circular economy goals.

Using AI for Dynamic Resource Allocation and Performance Optimization in Software Systems

This research work aims at evaluating the possibilities of using artificial intelligence in dynamic resource management and optimization of software system performance. In today’s complex world of application usage, normal methods of resource management are unable to cater to these dynamic needs and fulfill its usage potential. In this work, an assessment of three mainstream AI techniques – reinforcement learning, neural network, and genetic algorithm – is performed based on performance indicators such as resource utilization and consumption, average response time, throughput, costs, prediction capability, stability, and time taken to converge. The results show that the neural networks have the best resource acquisition performance as well as response rates, while the reinforcement learning has the best cost management and flexibility rates. As it has been pointed out, genetic algorithms are quite useful in finding optimization solutions, however real-time responsiveness is lack. Thus, the results provide significant understating of how to choose the proper AI technique depending on the specific application needs which in turn will be useful for organizations willing to improve their resource management using AI-based solutions.

Decentralized Machine Learning for Dynamic Resource Optimization in Wireless Networks using Reinforcement Learning

Efficient allocation of resources is crucial for optimizing wireless networks that face constraints in bandwidth, power, and spectrum. This paper proposes a decentralized reinforcement learning (RL) model that departs from traditional centralized paradigms to revolutionize resource optimization. The proposed model empowers individual wireless devices with autonomous decision-making capabilities, enhancing adaptability and scalability by leveraging Deep Q-Network (DQN) and Proximal Policy Optimization (PPO). The innovative integration of memory mechanisms facilitates learning from past experiences, addressing the dynamic nature of wireless environments. This decentralized RL model offers practical implications for improved efficiency, adaptability, and reliability in wireless network resource optimization. By transforming individual devices into collaborative decision-makers, our proposed model contributes to a resilient and responsive wireless communication infrastructure. The specific contributions of this paper include the pioneering use of DQN and PPO algorithms within a multi-agent system, offering a groundbreaking solution for dynamic resource optimization in wireless networks.

DRL-Based Dynamic Resource Configuration and Optimization for B5G Network Slicing

DRL-Based Dynamic Resource Configuration and Optimization for B5G Network Slicing

Computation Resource Optimization for Large-Scale Intelligent Urban Rail Transit: A Mean-Field Game Approach

Offloading tasks in smart devices (SDs) to an edge intelligence-empowered service centre (EISC) is a promising solution to support burgeoning intelligent applications in large-scale intelligent urban rail transits (URTs). However, dynamic computation resource allocation is still a crucial challenge, facing that various large-scale SDs share the EISC computation resources and the reality that the allocated computation resource for an SD is coupled with the offloading rate of all SDs. This paper proposes a joint dynamic offloading rate control and computation resource optimization method for large-scale intelligent URTs. Firstly, we model the large-scale multi-agent computation resource competition problem by a multi-player differential game (MPDG) and prove that the Nash equilibrium (NE) based optimal solution exists for each SDs. Then, we transform the MPDG model into a mean-field game (MFG). By introducing the mean-field into the game, we can solve the multi-agent optimization problem with a single-agent optimization method. We illustrate the rationality of the MFG model and propose an iterative solution method based on the finite difference method to derive the solution. Finally, we propose a z-transforming-based control method to dynamically reschedule computation resources among intelligent applications to achieve a satisfactory quality of service (QoS). Extensive simulation results show that our proposed scheme can significantly improve the performance of intelligent URTs.

Wireless Network Design Optimization for Computer Teaching with Deep Reinforcement Learning Application

ABSTRACT Computer technology has had a significant impact on the field of education, and its use in classrooms has facilitated the spread of knowledge and has helped students become well-rounded citizens. However, as the number of users accessing wireless networks for computer-based learning continues to grow, the scarcity of spectrum resources has become more apparent, which makes it crucial to find intelligent methods to improve spectrum utilization in wireless networks. Dynamic spectrum access is a critical technology in wireless networks, and it primarily focuses on how users can efficiently access licensed spectrum in a dynamic environment. This technology is a crucial means to tackle the problem of spectrum scarcity and low spectrum utilization. This work proposes a novel approach to address the issue of dynamic channel access optimization in wireless networks and investigates the dynamic resource optimization problem with deep reinforcement learning algorithms. The proposed approach focuses on the optimization of dynamic multi-channel access under multi-user scenarios and considers the collision and interference caused by multiple users accessing a channel simultaneously. Each user selects a channel to access and transmit data, and the network aims to develop a multi-user strategy that maximizes network benefits without requiring online coordination or information exchange between users. What makes this work unique is its utilization of deep reinforcement learning algorithms and a Long Short-Term Memory (LSTM) network that keeps an internal state and combines observations over events. This approach allows the network to utilize the history of processes to estimate the true state, providing valuable insights into how deep reinforcement learning algorithms can be used to optimize dynamic channel access in wireless networks. The work’s contribution lies in demonstrating that this approach is an effective means to solve the dynamic resource optimization problem, enabling the development of a multi-user strategy that maximizes network benefits. This approach is particularly valuable as it does not require online coordination or information exchange between users, which can be challenging in real-world scenarios. The proposed approach presents an important contribution to the field of dynamic spectrum access and wireless network optimization. As the demand for computer-based learning continues to increase, the use of intelligent methods to improve spectrum utilization in wireless networks will become even more critical. The findings of this work could have significant implications for the future of computer-based learning and education, enabling more efficient use of wireless networks and the creation of well-rounded citizens.

WiFi6 Dynamic Channel Optimization Method for Fault Tolerance in Power Communication Network

As the scale of power networks has expanded, the demand for multi-service transmission has gradually increased. The emergence of WiFi6 has improved the transmission efficiency and resource utilization of wireless networks. However, it still cannot cope with situations such as wireless access point (AP) failure. To solve this problem, this paper combines orthogonal frequency division multiple access (OFDMA) technology and dynamic channel optimization technology to design a fault-tolerant WiFi6 dynamic resource optimization method for achieving high quality wireless services in a wirelessly covered network even when an AP fails. First, under the premise of AP layout with strong coverage over the whole area, a faulty AP determination method based on beacon frames (BF) is designed. Then, the maximum signal-to-interference ratio (SINR) is used as the principle to select AP reconnection for the affected users. Finally, this paper designs a dynamic access selection model (DASM) for service frames of power Internet of Things (IoTs) and a scheduling access optimization model (SAO-MF) based on multi-frame transmission, which enables access optimization for differentiated services. For the above mechanisms, a heuristic resource allocation algorithm is proposed in SAO-MF. Simulation results show that the method can reduce the delay by 15% and improve the throughput by 55%, ensuring high-quality communication in power wireless networks.

Learning-based Phase-aware Multi-core CPU Workload Forecasting

Predicting workload behavior during workload execution is essential for dynamic resource optimization in multi-processor systems. Recent studies have proposed advanced machine learning techniques for dynamic workload prediction. Workload prediction can be cast as a time series forecasting problem. However, traditional forecasting models struggle to predict abrupt workload changes. These changes occur because workloads are known to go through phases. Prior work has investigated machine-learning-based approaches for phase detection and prediction, but such approaches have not been studied in the context of dynamic workload forecasting. In this article, we propose phase-aware CPU workload forecasting as a novel approach that applies long-term phase prediction to improve the accuracy of short-term workload forecasting. Phase-aware forecasting requires machine learning models for phase classification, phase prediction, and phase-based forecasting that have not been explored in this combination before. Furthermore, existing prediction approaches have only been studied in single-core settings. This work explores phase-aware workload forecasting with multi-threaded workloads running on multi-core systems. We propose different multi-core settings differentiated by the number of cores they access and whether they produce specialized or global outputs per core. We study various advanced machine learning models for phase classification, phase prediction, and phase-based forecasting in isolation and different combinations for each setting. We apply our approach to forecasting of multi-threaded Parsec and SPEC workloads running on an eight-core Intel Core-i9 platform. Our results show that combining GMM clustering with LSTMs for phase prediction and phase-based forecasting yields the best phase-aware forecasting results. An approach that uses specialized models per core achieves an average error of 23% with up to 22% improvement in prediction accuracy compared to a phase-unaware setup.

Predictive decision and reliable accessing for UAV communication in space-air-ground integrated networks

The cooperation of multiple Unmanned Aerial Vehicles (UAVs) has become a promising scenario in Space-Air-Ground Integrated Networks (SA-GINs) recently due to their widespread applications, where wireless communication is a basic necessity and is normally categorized into control and non-payload communication (CNPC) as well as payload communication. In this paper, we attempt to tackle two challenges of UAV communication respectively on establishing reliable CNPC links against the high mobility of UAVs as well as changeable communication conditions, and on offering dynamic resource optimization for Quality-of-Service (QoS) guaranteed payload communication with variable link connectivity. Firstly, we propose the concept of air controlling center (ACC), a virtual application equipped on the infrastructure in SAGINs, which can collect global information for estimating UAV trajectory and communication channels. We then introduce the knapsack problem for modelling resource optimization of UAV communication in order to provide optimal access points for both CNPC and payload communication. Meanwhile, using the air controlling information, predictive decision algorithm and handover strategy are introduced for the reliable connection with multiple access points. Simulation results demonstrate that our proposal ensures an approximate always-on reliable accessing of communication links and outperforms the existing methods against high mobility, sparse distribution, and physical obstacles.

Service-Oriented Dynamic Resource Slicing and Optimization for Space-Air-Ground Integrated Vehicular Networks

In this paper, we study Space-Air-Ground integrated Vehicular Network (SAGVN), and propose an online control framework to dynamically slice the SAG spectrum resource for isolated vehicular services provisioning. In particular, at a given time slot, the system makes online decisions on the request admission and scheduling, UAV dispatching, and resource slicing for different services. To characterize the impact of those parameters, we construct a time-averaged queue stability criteria by taking queue backlogs of all services into consideration, and formulate a system revenue function which incorporates the time-averaged system throughput and UAV dispatching cost. The objective is to maximize the system revenue while stabilizing the time-averaged queue, which falls into the scope of Lyapunov optimization theory. By bounding the drift-plus-penalty, the original problem can be decoupled into four independent subproblems, each of which is readily solved. The merits of our control framework are three-fold: 1) the system is able to admit and process as many requests as possible (i.e., maximizing the time-averaged throughput); 2) the time-averaged UAV dispatching cost is minimized; and 3) service queues are stabilized in the long-term. Extensive simulations are carried out, and the results demonstrate that the control framework can effectively achieve the system revenue maximization and queueing stabilization. Moreover, it can balance the trade-off among system throughput, UAV dispatching cost, and queueing states via parameter tuning. Compared with the fixed slicing, our dynamic slicing can react to the vehicular environment rapidly and achieve an average 26% of throughput improvement.

Dynamic Resource Optimization for Decentralized Estimation in Energy Harvesting IoT Networks

We study decentralized estimation of time-varying signals at a fusion center, when energy harvesting sensors transmit sampled data over rate-constrained links. We propose dynamic strategies to select radio parameters, sampling set, and harvested energy at each node, with the aim of estimating a time-varying signal while ensuring: 1) the accuracy of the recovery procedure and 2) the stability of the batteries around a prescribed operating level. The approach is based on stochastic optimization tools, which enable adaptive optimization without the need of apriori knowledge of the statistics of radio channels and energy arrivals processes. Numerical results validate the proposed approach for decentralized signal estimation under communication and energy constraints typical of Internet-of-Things scenarios.

A Hierarchical Approach to Resource Allocation in Extensible Multi-Layer LEO-MSS

Low earth orbit mobile satellite system (LEO-MSS) is the major system to provide communication support for mobile terminals beyond the coverage of terrestrial communication systems. However, the quick movement of LEO satellites and current single-layer system architecture impose restrictions on the capability to provide satisfactory service quality, especially for the remote and non-land regions with high traffic requirement. To tackle this problem, high-altitude platforms (HAPs) and terrestrial relays (TRs) are introduced to cover hot-spot regions, and the current single-layer system becomes an LEO-HAP multi-layer access network. Under this setup, we propose a hierarchical resource allocation approach to circumvent the complex management caused by the intricate relationships among different layers. Specifically, to maximize the throughputs, we propose a dynamic multi-beam joint resource optimization method in LEO-ground downlinks based on the predicted movement of LEO satellites. Afterwards, we propose the dynamic resource optimization method of HAP-ground downlinks when LEO satellites and HAPs share the same spectrum. To solve these problems, we use the Lagrange dual method and Karush-Kuhn-Tucker (KKT) conditions to find the optimal solutions. Numerical results show that the proposed architecture outperforms current LEO-MSS in terms of average capacity. In addition, the proposed optimization methods increase the throughputs of LEO-ground downlinks and HAP-ground downlinks with an acceptable complexity.

Capacity and Frequency Optimization of Wireless Backhaul Network Using Traffic Forecasting

Telecom operators are aiming to provide high-grade data, multimedia applications and low latency videos for smart devices. As today’s mobile data is experiencing rapid growth and the usage of smart devices are fabricating unparalleled challenges for telecom operators to meet the global bandwidth requirement. From the first generation to fourth generation, the technology evolution is predominantly governed by the hardware side but now it is moving towards the concept of cognitive network management, resource orchestration and machine learning-based solutions. In this paper, we propose the adaptive capacity and frequency optimization (ACFO) method for adaptive optimization based on time series forecasting approach. The daily capacity utilization of microwave (MW) links is analyzed to use forecasted demand. Based on the projected demand, the capacity and frequency optimization will be executed. The two main forecasting models 1) SARIMA and 2) MLP are used and for performance evaluation, we used RMSE and MAPE criterion. The analytic outcomes show that MLP with two layers and six hidden nodes (6/6) are good enough to achieve the desired results. In some cases, we need to exceed the hidden nodes up to fifteen (15/15). By using the forecasting approach, the reactive optimization will successfully shift to the predicted/proactive optimization, will balance the resource distribution and can condense the wastage of resources. The outcome of the study will be a contribution to the dynamic resource optimization in wireless backhaul network.

Buffer-Aware Streaming in Small-Scale Wireless Networks: A Deep Reinforcement Learning Approach

Buffer-aware video streaming, which exploits the available storage space in user device to store the prefetched video data in good channels for video data use in poor channels, has been proved to have the potential to reduce the impact of fluctuating wireless channels on user-perceived video performance. However, in practical wireless networks, due to the unknown channel state and video rate, providing buffer-aware video streaming service to wireless user is a challenging problem. In this paper, with the aim to design an autonomous wireless video streaming system, we apply the deep reinforcement learning approach to dynamic resource optimization for wireless buffer-aware video streaming under unknown channel state and video rate. Specifically, we define a reward function for buffer-aware video streaming as the effective video streaming time when neither video-playback overflow nor video-playback underflow occurs. We propose a Markov decision process based problem formulation of the joint bandwidth allocation and buffer management for maximizing the effective video streaming time of all users. The optimal bandwidth allocation and buffer management policy is learned from training a deep neural network based on a deep reinforcement learning algorithm. We simulate the proposed algorithm in Tensorflow. Simulation results verify that the proposed deep reinforcement learning approach is effective for buffer-aware video streaming in wireless networks.

Moving Forward: A Simulation-Based Approach for Solving Dynamic Resource Management Problems

Standard dynamic resource optimization approaches, such as value function iteration, are challenged by problems involving complex uncertainty and a large state space. We extend a solution technique to address these limitations called approximate dynamic programming (ADP). ADP recently emerged in the macroeconomics literature and is novel to bioeconomics. We demonstrate ADP in solving a simple fishery management model under uncertainty to show: the mechanics of ADP in simplest form; the accuracy of ADP; the value of a nonparametric extension; and readily adaptable, non-specialized code. We then demonstrate ADP’s capacity to handle rich bioeconomic problems by solving the fishery management problem subject to four autocorrelated shock processes (governing economic returns and biological dynamics) which entails four sources of stochasticity and five continuous state variables. We find that accounting for multiple autocorrelation has important impacts on harvest policy and generates gains that depend crucially on the structure of harvest cost.

A Dynamic Multi-Layer Resource Allocation and Optimization Framework in Application-Centric Networks

In a software-defined networking (SDN) based network, service requests can be accommodated according to the network application requirements. We devise a framework where such requirements drive IP and optical network resource allocation, dynamic optimization, and instantiation through an SDN orchestrator. The framework aims to balance the often-conflicting goals such as energy savings and resource allocation speed, while ensuring that the orchestrator can accommodate incoming service requests without interrupting the running services. We show the functionality of the framework with a simulation on a real network topology and a realistic input traffic.

Sleep-Wake Cycling and Energy Conservation: Role of Hypocretin and the Lateral Hypothalamus in Dynamic State-Dependent Resource Optimization.

The hypocretin (Hcrt) system has been implicated in a wide range of physiological functions from sleep-wake regulation to cardiovascular, behavioral, metabolic, and thermoregulagtory control. These wide-ranging physiological effects have challenged the identification of a parsimonious function for Hcrt. A compelling hypothesis suggests that Hcrt plays a role in the integration of sleep-wake neurophysiology with energy metabolism. For example, Hcrt neurons promote waking and feeding, but are also sensors of energy balance. Loss of Hcrt function leads to an increase in REM sleep propensity, but a potential role for Hcrt linking energy balance with REM sleep expression has not been addressed. Here we examine a potential role for Hcrt and the lateral hypothalamus (LH) in state-dependent resource allocation as a means of optimizing resource utilization and, as a result, energy conservation. We review the energy allocation hypothesis of sleep and how state-dependent metabolic partitioning may contribute toward energy conservation, but with additional examination of how the loss of thermoregulatory function during REM sleep may impact resource optimization. Optimization of energy expenditures at the whole organism level necessitates a top-down network responsible for coordinating metabolic operations in a state-dependent manner across organ systems. In this context, we then specifically examine the potential role of the LH in regulating this output control, including the contribution from both Hcrt and melanin concentrating hormone (MCH) neurons among a diverse LH cell population. We propose that this hypothalamic integration system is responsible for global shifts in state-dependent resource allocations, ultimately promoting resource optimization and an energy conservation function of sleep-wake cycling.

Energy Efficient Dynamic Resource Optimization in NOMA System

Non-orthogonal multiple access (NOMA) with successive interference cancellation (SIC) is a promising technique for next generation wireless communications. Using NOMA, more than one user can access the same frequency-time resource simultaneously and multi-user signals can be separated successfully using SIC. In this paper, resource allocation algorithms for subchannel assignment and power allocation for a downlink NOMA network are investigated. Different from the existing works, here, energy efficient dynamic power allocation in NOMA networks is investigated. This problem is explored using the Lyapunov optimization method by considering the constraints on minimum user quality of service and the maximum transmit power limit. Based on the framework of Lyapunov optimization, the problem of energy efficient optimization can be broken down into three subproblems, two of which are linear and the rest can be solved by introducing a Lagrangian function. The mathematical analysis and simulation results confirm that the proposed scheme can achieve a significant utility performance gain and the energy efficiency and delay tradeoff is derived as [O(1/V), O(V)] with V as a control parameter under maintaining the queue stability.