Dynamic Resource Allocation in Cloud Networks Using Deep Learning : A review

Teaching resources in higher education are an important part of the talent cultivation system of higher education institutions. Dynamic management and optimal allocation of teaching resources are the focus of the construction of each school, which reflects the teaching, research, and management level of the school to a certain extent. In this paper, after combining the current situation and management problems of teaching resources in six schools in City A and putting forward constructive opinions, we further use data envelopment analysis to study the efficiency of teaching resource allocation in School H in City A and put forward the method of resource optimization.The input redundancy rate of School X1 in School H during the three years of 2021-2023 is 30.25%, 44.15%, and 55.26%, respectively. The data shows that the school has an excess of teacher inputs and there is a shortage of resources for teachers. For the current dynamic management and resource allocation problems of teaching resources in A city school, this paper argues that the management of school teaching resources is a long-term and complex organic whole, and needs to be reformed in terms of structure, resource integration and management, and to construct a high-quality and efficient management model. In terms of resource optimization and allocation, schools can optimize the talent resource allocation mechanism and other aspects to promote the full allocation of teaching resources in higher education.

The advent of sixth-generation (6G) mobile networks introduces unprecedented demands for ultra-low latency, terabit-per-second throughput, and dynamic resource allocation across heterogeneous services such as URLLC, eMBB, and mMTC. Traditional resource management approaches lack the adaptability and predictive intelligence required to operate under such conditions. This study proposes a comparative analysis of deep learning models—Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Transformer architectures—for intelligent resource allocation in 6G environments. Synthetic traffic patterns and benchmark datasets were used to emulate realistic 6G workloads, and models were evaluated using key performance metrics: accuracy, latency, throughput, energy efficiency, and fairness index. Simulations were conducted in Python using TensorFlow and PyTorch frameworks within SDN-based virtualized environments to replicate programmable 6G infrastructures. The proposed hybrid system, integrating CNN, RNN, and Transformer modules, achieved latency reduction of 38%, throughput improvement of 22%, energy savings of 26%, and fairness index gain of 18% compared to baseline models. These findings demonstrate the viability of deep learning for scalable, adaptive, and sustainable resource allocation in next-generation mobile networks. The study contributes a reproducible simulation framework, a performance benchmark across models, and a foundation for future research in federated learning, edge AI, and 6G deployments.

Purpose: The general objective of the study is to assess the influence of strategic planning and service delivery at Secondary schools in Kigali-Rwanda. The study is guided by the following specific objectives: To assess the influence of direction setting on service delivery at Association des Parents Advertistes Pour le Development del’Education, to identify the influence of resource allocation on service delivery at Association des Parents Advertistes Pour le Development del’Education and to evaluate the influence of organization structure on service delivery at Association des Parents Advertistes Pour le Development del’Education. This study is significant in the following ways; the findings of this study show extent in which strategic planning are applied as well as its contribution on the service delivered by secondary schools. The study helps the researcher to gain knowledge of conducting research through the first-hand experience of data collection, analysis, and interpretation specifically on the influence of strategic planning on the service delivery at Secondary Schools in Kigali, Rwanda. Methodology: This study adopted descriptive research design to get results related to the study, the target population as well as sample size was 143. Researcher applied census sampling. Findings: The key findings of the study indicate a strong and significant relationship between the independent variables’ direction setting, resource allocation, and organizational structure and service delivery at the Association des Parents Adventistes Pour le Développement de l’Éducation. The multiple regression model revealed a high correlation (R = 0.956) and explained approximately 91.5% of the variance in service delivery (R² = 0.915, Adjusted R² = 0.913), indicating an excellent model fit. All three predictors were statistically significant: direction setting (β = 0.212, p = 0.004), resource allocation (β = 0.128, p = 0.002), and organization structure (β = 0.651, p < 0.001), suggesting that these factors positively influence service delivery. Organizational structure had the strongest effect, followed by direction setting and resource allocation, highlighting its critical role in enhancing coordination, communication, and overall effectiveness of service delivery in the school. the study concluded that the effectiveness of service delivery at APADE is strongly influenced by direction setting, resource allocation, and organizational structure. Unique Contribution to Theory, Practice and Policy: The study concludes that direction setting, resource allocation, and organizational structure significantly influence service delivery at the Association des Parents Adventistes Pour le Développement de l’Éducation, with organizational structure having the strongest impact. It is recommended that the school continually clarify its vision and strategic goals, optimize resource allocation, and maintain an effective organizational structure to enhance coordination, communication, and overall service delivery. For future research, studies should explore additional factors such as stakeholder engagement, cultural influences, and technological integration to further understand their effects on educational service delivery in secondary schools.

Background: Non-orthogonal multiple access (NOMA) is viewed as the key multiple access technology for 5G and beyond networks, attracting the attention of academics and industries. NOMA and the multiple input multiple output (MIMO-NOMA) technology can improve a system’s throughput, latency, and energy efficiency (EE) in future-generation communication networks. Objective: The objective of this paper is to achieve maximum EE by applying the Max-min Power Control Algorithm (MMPCA) through sub-channel optimization, resource allocation (RA) optimization, access point selection (APS), and user association. The EE results obtained with and without using MMPCA are compared to the RA optimization from a conventional water-filling algorithm (WFA). Method: This paper formulates a framework for user-centric (UC) joint resource allocation, such as backhaul connection via beam-forming and Access point (AP) to user connection via MIMO-NOMA. The user without interference is decoded using the NOMA principle. The MMPCA was also used to optimize cooperative power allocation, sub-channel allocation, and efficient user association. The RA for EE is framed as a mixed non-convex and non-linear function using successive convex approximation and sum ratio decoupling converted into convex and linear. A bisection method was used to achieve optimal RA, user association, and sub-channel assignment. Results and Conclusion: The simulation shows energy efficiency (EE) improvement. Similarly, it is observed that MMPCA outperforms the WFA.

In this letter, we investigate the adaptive resource allocation with Lyapunov optimization for time-varying wireless networks. The transmit power minimization is characterized by a stochastic optimization model. A dynamic resource allocation (DRA) algorithm is developed to accommodate the wireless network dynamics, i.e., time-varying wireless channels and as well the queuing dynamics. We investigate the tracking error between the DRA algorithm output and the target optimal resource allocation solution. Based on these results, we further develop an adaptive-compensation resource allocation (ACRA) algorithm, which iterates only once when the network state changes for saving the huge iteration overheads. Finally, we determine a sufficient condition that the ACRA algorithm asymptotically tracks the moving equilibrium point with no tracking errors. Simulation results validate the theoretical analysis of our proposed scheme.

In wide-area networks (WANs), efficiently allocating distributed resources presents a significant challenge due to increasing node numbers and complex network conditions. Traditional algorithms, which heavily depend on global information, encounter scalability limitations. This study proposes an adaptive distributed optimal resource allocation consistency algorithm designed for WANs. The approach dynamically adjusts resource allocation based on real-time network conditions and user requirements, reducing dependence on global data. A time window distribution model and an information coding model are employed to optimize the resource allocation process. By integrating a Q-learning algorithm and extending the Paxos algorithm to WANs, global consistency among network nodes is ensured. The experimental results demonstrate the algorithm’s superior performance, particularly in low-bandwidth WAN scenarios (below 2 Mb/s), where it significantly enhances throughput. An average resource utilization rate exceeding 97% is achieved, along with rapid convergence and high parallelism. This research underscores the algorithm’s effectiveness and scalability, providing a robust solution to distributed resource allocation challenges in complex WAN environments.

Edge video analytics enables agile responses of machine-centric applications by streaming videos from end devices to edge servers (ESs) for resource-intensive deep neural network (DNN) inference. Quality of Inference (QoI), reflected by end-to-end analytics delay and inference accuracy, is crucial for the high-quality real-time decision-making. Due to the limited ES resources and dynamic network conditions, video configuration adaptation and fine-grained resource allocation are necessary to enhance the QoI of video streams, which involves a delicate balance between delay and accuracy. In this article, we propose an edge-enabled multivideo analytics framework, in which performing DNN inference relies on heterogeneous resources, including CPU, GPU, and memory. Considering the different impacts of heterogeneous resources on QoI and double-queue backlogs, we formulate the joint video configuration adaptation, CPU-GPU resource allocation, and batch size selection (JCCGB) problem based on semi-Markov decision process to maximize the long-term average QoI. Then, Lyapunov optimization is applied to transform the original problem into one that minimizes the Lyapunov drift-plus-penalty upper bound, thereby ensuring the stability of both the transmission and computation queues. To tackle the potential multimodality of the optimal configuration adaptation and resource allocation policy, we propose the diffusion-deep-reinforcement-learning-based JCCGB (DD-JCCGB) scheme. This approach effectively enhances the policy performance and accelerates the training process. Experiments driven by real-world network traces demonstrate that the DD-JCCGB algorithm improves the long-term average QoI and outperforms baseline schemes.

In this paper, we provide simulation results for the optimal channel-aware adaptive transmission, dynamic resource allocation and multiuser scheduling schemes for the downlink of a wireless cell that is assisted by a number of relays. The relays are half-duplex and apply a regenerative decode-and-forward scheme, and the channels are Rayleigh block-faded. We apply the optimal resource allocation scheme for two specific scheduling criteria, the maximum capacity scheduler and the proportional fair scheduler. For both scheduling schemes, we study the gains of using relays on the overall aggregate throughput of the system as well as for the individual users. Furthermore, we compare the performance versus the total number of relays and their locations in the wireless cell and we study the statistics of relays' usage. Moreover, we compare between the two scheduling schemes in terms of achievable rate and fairness among users.

In cluster-based cognitive radio networks, secondary system uses available channels that are not used temporally by primary systems. In multi-channel operational environments, each available channel may have different wireless channel gain and primary activity so that achievable data rate to secondary users (SUs) and required sensing parameter value are channel dependent. SUs also have different energy saving requirements and data traffic demands. Therefore, based on measured channel conditions and user constraints, cluster head needs to decide which channel should be allocated to which SUs and to configure the optimum MAC frame structure for satisfying each SU’s service demands. Furthermore, optimization to provide proportional fairness among the cognitive secondary users in resource allocation in terms of energy consumption and data rate is very important. In this paper, a dynamic MAC frame configuration and optimal resource allocation scheme for multi-channel ad-hoc cognitive radio network is proposed. We formulate our dynamic resource allocation model as a constrained optimization problem with multi-objective functions using particle swarm optimization (PSO) algorithm. The proposed PSO scheme guarantees that the allocation captures the individual traffic and energy saving demands and maximizes the objectives functions simultaneously. Simulation results show the proposed scheme can successfully maximize the intended utility function and provide proportional fairness between SUs.

System identification extracts information from a system's operational data to derive a representative model for the system. Studies of system identification have been concentrated on estimation algorithms and their convergence. Focusing on optimal resource allocation under a given reliability requirement, this paper studies identification complexity and its relations to decision making. Dynamic resource assignments are investigated. Resource allocation algorithms are developed and their convergence properties are established. Illustrative examples demonstrate better resource management than worst-case strategies when our algorithms are applied.

In this paper, we present and evaluate the performance of a resource allocation algorithm to enhance the Quality of Service (QoS) provision and energy efficiency of uplink Long Term Evolution (LTE) systems. The proposed algorithm considers the main constraints in uplink LTE resource allocation, i.e., the allocation of contiguous sets of resource blocks of the localized Single Carrier—Frequency Division Multiple Access (SC-FDMA) physical layer to each user, and the imperfect knowledge of the users' uplink buffer status and packet waiting time. The optimal resource allocation is formulated as a discrete connected cake-cutting problem, where different agents are allocated consecutive subsequences of a sequence of indivisible items. This problem is NP-hard, therefore a suboptimal algorithm is introduced, which performs resource allocation using information on the estimated uplink packet delay, the average delay and data rate of past allocations, as well as the required uplink power per resource block. Based on simulation results, the proposed algorithm achieves significant performance improvement in terms of packet timeout rate, goodput, and fairness. Moreover, the effect of poor QoS provision on energy efficiency is demonstrated through the evaluation of the performance in terms of energy consumption per successfully received bit.

This paper proposes a dynamic resource allocation scheme to maximize the energy efficiency (EE) for Massive MIMO Systems. The imperfect channel estimation (CE) and feedback are explicitly considered in the EE maximization problem, which aim to optimize the power allocation, the antenna subset selection for transmission, and the pilot assignment. Assuming CE error to be bounded for the complex-constrained Cramer–Rao Bound (CRB), theoretical results show that the lower bound is directly proportional to its number of unconstrained parameters. Utilizing this perspective, a separated and bi-directional estimation is developed to achieve both low CRB and low complexity by exploiting channel and noise spatial separation. Exploiting global optimization procedure, the optimal resource allocation can be transformed into a standard convex optimization problem. This allows us to derive an efficient iterative algorithm for obtaining the optimal solution. Numerical results are provided to demonstrate that the outperformance of the proposed algorithms are superior to existing schemes.

Energy efficiency is an important design criterion for wireless\ncommunications. When parallel frequency channels are used for multiuser\ntransmission, the channel bandwidths and user power are adjusted to maximize\nthe sum information rate with the bandwidth budget, the transmit power budget,\nand the user-specific rate requirements. The maximum sum rate is used in\nmeasuring the energy efficiency. With fixed or flexible bandwidths of the\nfrequency channels, practical methods are developed to find the total transmit\npower with the unique optimal resource (bandwidth and power) allocation for\nmaximum energy efficiency. This resource allocation ensures that, while each\nuser's minimum rate requirement is satisfied, all the excess resource of the\nspectrum and transmit power is dedicated to the one user with the best channel\nquality. Simulation results validate the optimal solutions of total transmit\npower and resource allocation that support the energy-efficient multiuser\ntransmission.\n

Unmanned aerial vehicles (UAVs) have been conceived as an available solution to substitute terrestrial base stations (TBSs) to provide downloading services for user equipment (i.e. mobile devices) that have difficulty communicating directly with TBSs. However, the mobility of user equipment (UE) and the random nature of the number of UE will cause several challenges including 1) the hardness of determining optimal user association and UAV resource (i.e., bandwidth and transmit power) allocation decision, 2) the burden of network function maintenance owing to the necessity of shutting down the entire system. Therefore, in this article, joint user association and resource allocation are designed for a software-defined network (SDN)-adopted leaderless softwarized UAV network, where each UAV is regarded as a flying SDN controller to enhance the control ability of the considered network. The purpose is to maximize energy efficiency (EE) with satisfying the quality of service (QoS). To this end, a joint method based on hierarchical agglomerative clustering (HAGC) and multi-agent deep deterministic policy gradient (MADDPG) is proposed. Specifically, the HAGC approach is utilized to determine the optimal MDs association with UAVs. Afterward, MADDPG approach is leveraged to obtain the best policy for resource allocation, aiming to achieve the maximum EE. Finally, the effectiveness of the proposed method is confirmed by the evaluation results.

With the advent of gaining popularity in cloud computing, there is an increase in demand of resources among the heterogeneous workload types on cloud. Resource management is a key challenging problem that is faced by the cloud service providers, by achieving business goals and agreed level of service with the subscribers. This paper focuses on dynamic resource allocation with risk analysis by meeting service level agreements. Further proposed framework handles heterogeneous workload types by dynamic capacity planning with risk assessment to maximise the profit and resource utilisation on clouds. In addition to advanced resource reservation, SLA-based scheduling/rescheduling with risks involved in resource allocation is considered in the proposed model. The experimental results demonstrate that proposed framework maximises the resource utilisation and profit gain of the cloud service provider when evaluated against widely used static configuration strategy.