Game theoretic resource allocation in cloud computing

Mobile Edge Computing (MEC) is a promising computing paradigm, which can provide mobile devices with powerful computing capability and low delay by offloading resource-intensive tasks to the MEC Server. However, due to the varying wireless channel conditions and different task delay requirements, the resource allocation decision made by MEC Server cannot always achieve the lowest system energy consumption. In this paper, we consider resource allocation problem in a multi-user MEC system. The objective is to achieve minimum system energy consumption under delay constraint by suitably allocating communication and computing resources. We solve the communication resources (subcarrier) allocation problem through regression algorithm, make use of three regression algorithms to predict subcarrier allocation. And then calculate the computing resources (CPU frequency) allocation and system energy consumption. Simulation results show that machine learning can achieve near-optimal system energy consumption, where the system energy consumption of the ANN(Artificial Neural Network), Xgboost (Extreme Gradient Boosting), and RF(Random Forest) algorithms is 1.30%, 7.49%, and 23.29% higher than the optimal system energy consumption respectively. In the view of prediction loss, ANN and Xgboost perform better than RF.

The convergence of transformative technologies, including the Internet of Things (IoT), Big Data, and Artificial Intelligence (AI), has driven private edge cloud systems to the forefront of research efforts. The access to massive terminals and the emergence of personalized services pose serious challenges for efficient resource management in power private edge cloud systems. To address the challenge of inequitable resource allocation in the private edge cloud, this work proposes an intelligent resource allocation strategy with a slicing and auction approach. By formalizing the resource allocation problem as a Mixed Integer Nonlinear Programming (MINLP) puzzle, the method transforms it into a hierarchical allocation challenge for Mobile Network Operators (MNOs), Mobile Virtual Network Operators (MVNOs), and power terminals. The proposed Multi-hop Progressive Auction Algorithm (MPAA) addresses the sliced resource allocation problem between MNOs and MVNOs. Furthermore, a Terminal Resource Allocation Strategy (TRAS) based on improved particle swarm optimization is proposed to solve the spectrum resource allocation problem between MVNOs and power terminals. Extensive simulation results show that the bidding overhead of MPAA is reduced by 6.12% and the average terminal satisfaction of TRAS is improved by about 1.3% compared to conventional methods, thus improving the utilization of wireless resources within the power AIoT.

Software testing Resource allocation and release time decisions are vital for the software systems. The objective behind such critical decisions may differ from firm to firm. The motive of the firm may be maximization of software reliability or maximization of number of faults to be removed from each module or it may be minimization of number of faults remaining in the software or minimization of testing resources. Taking into consideration these different aims, various authors have investigated the problem of resource allocation and release time problem. In this paper we investigate various software release policies and resource allocation problem, for example, policies based on the dual constraints of cost and reliability.

Automatic decision-making approaches, such as reinforcement learning (RL), have been applied to (partially) solve the resource allocation problem adaptively in the cloud computing system. However, a complete cloud resource allocation framework exhibits high dimensions in state and action spaces, which prohibit the usefulness of traditional RL techniques. In addition, high power consumption has become one of the critical concerns in design and control of cloud computing systems, which degrades system reliability and increases cooling cost. An effective dynamic power management (DPM) policy should minimize power consumption while maintaining performance degradation within an acceptable level. Thus, a joint virtual machine (VM) resource allocation and power management framework is critical to the overall cloud computing system. Moreover, novel solution framework is necessary to address the even higher dimensions in state and action spaces. In this paper, we propose a novel hierarchical framework for solving the overall resource allocation and power management problem in cloud computing systems. The proposed hierarchical framework comprises a global tier for VM resource allocation to the servers and a local tier for distributed power management of local servers. The emerging deep reinforcement learning (DRL) technique, which can deal with complicated control problems with large state space, is adopted to solve the global tier problem. Furthermore, an autoencoder and a novel weight sharing structure are adopted to handle the high-dimensional state space and accelerate the convergence speed. On the other hand, the local tier of distributed server power managements comprises an LSTM based workload predictor and a model-free RL based power manager, operating in a distributed manner.

The problem of joint radio and cloud resources allocation is studied for heterogeneous mobile cloud computing networks. The objective of the proposed joint resource allocation schemes is to maximize the total utility of users as well as satisfy the required quality of service (QoS) such as the end-to-end response latency experienced by each user. We formulate the problem of joint resource allocation as a combinatorial optimization problem. Three evolutionary approaches are considered to solve the problem: genetic algorithm (GA), ant colony optimization with genetic algorithm (ACO-GA), and quantum genetic algorithm (QGA). To decrease the time complexity, we propose a mapping process between the resource allocation matrix and the chromosome of GA, ACO-GA, and QGA, search the available radio and cloud resource pairs based on the resource availability matrixes for ACO-GA, and encode the difference value between the allocated resources and the minimum resource requirement for QGA. Extensive simulation results show that our proposed methods greatly outperform the existing algorithms in terms of running time, the accuracy of final results, the total utility, resource utilization and the end-to-end response latency guaranteeing.

Multiuser Orthogonal Frequency Division Multiple Access (OFDMA) networks, such as Long Term Evolution networks, use the frequency reuse-1 model to face the tremendous increase of mobile traffic demands, and to increase network capacity. However, inter-cell interference problems are generated, and they have a negative impact on cell-edge users performance. Resource and power allocation should be managed in a manner that alleviates the negative impact of inter-cell interference on system performance. In this paper, we formulate a novel centralized multi-cell resource and power allocation problem for multiuser OFDMA networks. The objective is to maximize system throughput while guaranteeing a proportional fair rate for all the users. We decompose the joint problem into two independent problems: a resource allocation problem and a power allocation problem. We prove that each of these problems is a convex optimization problem, and that their optimal solution is also an optimal solution to the original joint problem. Lagrange duality theory and subgradient projection method are used to solve the centralized power allocation problem. We study the convergence of our centralized approach, and we find out that it reduces inter-cell interference, and increases system throughput and spectral efficiency in comparison with the frequency reuse-1 model, reuse-3 model, fractional frequency reuse, and soft frequency reuse techniques.

The QoS based Resource Allocation Model (Q-RAM) proposed by R. Rajkumar et al. (1998) presented an analytical approach for satisfying multiple quality of service dimensions in a resource constrained environment. Using this model, available system resources can be apportioned across multiple applications such that the net utility that accrues to the end users of those applications is maximized. We present several practical solutions to allocation problems that were beyond the limited scope of Q-RAM. We show that the Q-RAM problem of finding the optimal resource allocation to satisfy multiple QoS dimensions is NP hard. We then present a polynomial solution for this resource allocation problem which yields a solution within a provably fixed and short distance from the optimal allocation. Secondly, Q-RAM dealt mainly with the problem of apportioning a single resource to satisfy multiple QoS dimensions. We study the converse problem of apportioning multiple resources to satisfy a single QoS dimension. In practice, this problem becomes complicated, since a single QoS dimension perceived by the user can be satisfied using different combinations of available resources. We show that this problem can be formulated as a mixed integer programming problem that can be solved efficiently to yield an optimal resource allocation. We also present the run times of these optimizations to illustrate how these solutions can be applied in practice. A good understanding of these solutions will yield insights into the general problem of apportioning multiple resources to satisfy simultaneously multiple QoS dimensions of multiple concurrent applications.

Spectrum access system (SAS) is a spectrum sharing framework proposed to share the spectrum between the incumbent users and the citizen broadband radio service devices, i.e. Priority access users and general authorized access (GAA) users. In this paper, we propose an interfering angle based method for the joint resource (channel and transmit power) allocation problem to the mobile and fixed GAA users. With mobile GAA users, the set of GAA users that can hear each other will change at different time instants making the resource allocation problem more challenging. The resource allocation of fixed and mobile GAA users is done considering coexistence with priority users, as well as coexistence between mobile and fixed GAA users. For the conflict-free resource allocation to fixed and mobile GAA users, we propose to use the maximum allowed transmit power for the beams of fixed GAA users that lie within the interference range of mobile GAA users. The simulation results show improved capacity from our proposed method while satisfying a predetermined interference constraint.

We study the problem of admission control and resource (subcarriers, power and bit-loading) allocation in a heterogeneous OFDMA wireless network serving both QoS-constrained high-priority users and best-effort users. By clustering subcarriers, efficient algorithms for cluster and power allocation have been proposed. Our strategy maximizes the total network utility of the best-effort users while satisfying the QoS request for maximum number of high-priority users. We approximate best-effort user utility function as a piece-wise linear function and propose a linear programming based cluster allocation algorithm. The feasibility of the resource allocation problem depends on the number of HP users in the network. Since a large number of demanding HP users would render the resource allocation problem infeasible, a joint admission control and resource allocation scheme could be an efficient way of tackling both problems with less overhead. By incorporating admission control, we propose an efficient and optimal joint admission control and resource allocation algorithm.

The subject matter of the article is the method of scheduling and active queues management on routers interfaces of telecommunication networks. The goal of the work is to develop and improve the method related to the optimization of queues management processes on routers interfaces of telecommunication networks and aimed at improving the indicators of the quality of service. The article solves optimization problems of Congestion Management, Resource Allocation and Congestion Avoidance on routers interfaces of telecommunication networks in the framework of the two-stage calculation method. Research methods are based on the results of analytical calculations obtained in the course of using the Optimization Toolbox package of the Mat lab environment. Wherein in the course of improving the queue management method on routers interfaces of telecommunication networks, a mathematical programming apparatus was used, containing linear programming functions. The following results were obtained – the improved method of scheduling and active queues management on routers interfaces of telecommunication networks with the coordinated, sequential and two-stage solution of such interface problems as Congestion Management, Resource Allocation, and Congestion Avoidance. Conclusions : using the proposed method of scheduling and active queue management had made to achieve an agreed solution to Congestion Management, Resource Allocation, and Congestion Avoidance problems. In the course of solving the optimization problem of Congestion Management, which was conditionally related to the first stage of calculations, it was possible to achieve optimal aggregation and allocation of packet flows over queues formed on the router interface based on the commensurability of their classes. In the course of solving the optimization problems of resource allocation, and congestion avoidance, which were conditionally related to the second stage of calculations, it was possible to achieve an optimal bandwidth allocation of the router interface between the queues formed on it, and minimize possible denials of service.

The access infrastructure to the Cloud is usually a major drawback that limits the uptake of Cloud services. Attention has turned to rethinking a new architectural deployment of the overall Cloud service delivery. We argue that it is not sufficient to integrate the cloud domain with the operator's network domain based on the current models. In this work we envision a full integration of the Cloud and the network, where cloud resources are no longer confined to a data center, but are spread throughout the network and owned by the network operator. In such an environment, challenges arise at different levels, such as at the resource management, where both cloud and network resources need to be managed in an integrated approach. We particularly address the resource allocation problem through joint virtualization of network and cloud resources, by proposing an algorithm to allocate cloud and network resources in an integrated way. This algorithm is evaluated through both simulation and experimental results in a real virtualization platform.

Recently, deep neural networks (DNNs) have been widely studied and achieved tremendous success in a variety of real-world applications, such as computer vision, natural language processing, medical diagnosis, and machine translation. Deep reinforcement learning (DRL), as an emerging powerful deep learning technique, combines DNNs with reinforcement learning into an interactive system that enables the agent to learn the best policy from the environment to optimize their rewards. DRL opens up many new applications in domains such as healthcare, robotics, and smart grids. With the rapid evolution of IT infrastructures, cloud computing has been witnessed as the most common computing paradigm. The underlying infrastructure of cloud computing relies on a large amount of data centers. Nowadays, with the fast development of communication technologies, such as 5G, cloud computing is now evolving into yet another new paradigm called edge computing, providing for emerging applications, especially in the Internet-of-Things. The energy efficiency issue from both "cloud" and "edge" are becoming more crucial and calls for more attention. To solve the real-world energy efficiency problems, we take advantage of the deep learning and deep reinforcement learning techniques for efficient modeling of both "edge" and "cloud" applications. On the "edge" side, wearable technologies and the Internet of Things (IoT) are becoming increasingly popular in both personal and commercial applications with the employment of the embedded system technologies. Non-Volatile Processors (NVPs) have been proposed for the intermittent power supply problem, to ensure the instant on/off for embedded systems under the unstable power supply. We proposed neural network-based prediction techniques to derive the best power extraction policy and converter parameters optimization technique for NVPs. In cloud computing, we formulated a novel optimal power management problem of data centers to minimize the total cost by proposing the long short term memory (LSTM) neural network-based method for the prediction of future data center power consumption, and further we presented a novel DRL-based hierarchical framework for solving the overall resource allocation and power management problem in cloud computing systems. Furthermore, we explored the effectiveness of the DRL-based framework for high dimensional action and state spaces problem-dynamic treatment regimes, to model real-life complexity in heterogeneous disease progression and treatment choices, to provide doctors and patients the data-driven personalized decision recommendations. Last but not least, we explored the efficiency of deep neural networks. Structured weight pruning is a representative model compression technique of DNNs to reduce both the storage and computation requirements and meanwhile accelerate inference. An automatic hyperparameter determination process is necessary due to a large number of flexible hyperparameters. We propose an automatic structured pruning framework to compress the DNNs. Our framework outperforms the prior works on automatic model compression by up to 33× in pruning rate (120× reduction in the actual parameter count) under the same accuracy. Significant inference speedup has been observed from the proposed framework on actual measurements on the smartphone.

The rapid growth of Industrial Internet of Things (IIoT) terminals has resulted in tasks exhibiting increased concurrency, heterogeneous resource demands, and dynamic priorities, significantly increasing the complexity of task scheduling in edge computing. Cloud-edge-end collaborative computing leverages cross-layer task offloading to alleviate edge node resource contention and improve task scheduling efficiency. However, existing methods generally neglect the joint optimization of task offloading, resource allocation, and priority adaptation, making it difficult to balance task execution and resource utilization under resource-constrained and competitive conditions. To address this, this paper proposes a two-stage dynamic-priority-aware joint task offloading and resource allocation method (DPTORA). In the first stage, an improved Multi-Agent Proximal Policy Optimization (MAPPO) algorithm integrated with a Priority-Gated Attention Module (PGAM) enhances the robustness and accuracy of offloading strategies under dynamic priorities; in the second stage, the resource allocation problem is formulated as a single-objective convex optimization task and solved globally using the Lagrangian dual method. Simulation results show that DPTORA significantly outperforms existing multi-agent reinforcement learning baselines in terms of task latency, energy consumption, and the task completion rate.

In a shared virtual computing environment, dynamic load changes as well as different quality requirements of applications in their lifetime give rise to dynamic and various capacity demands, which results in lower resource utilization and application quality using the existing static resource allocation. Furthermore, the total required capacities of all the hosted applications in current enterprise data centers, for example, Google, may surpass the capacities of the platform. In this paper, we argue that the existing techniques by turning on or off servers with the help of virtual machine (VM) migration is not enough. Instead, finding an optimized dynamic resource allocation method to solve the problem of on-demand resource provision for VMs is the key to improve the efficiency of data centers. However, the existing dynamic resource allocation methods only focus on either the local optimization within a server or central global optimization, limiting the efficiency of data centers. We propose a two-tiered on-demand resource allocation mechanism consisting of the local and global resource allocation with feedback to provide on-demand capacities to the concurrent applications. We model the on-demand resource allocation using optimization theory. Based on the proposed dynamic resource allocation mechanism and model, we propose a set of on-demand resource allocation algorithms. Our algorithms preferentially ensure performance of critical applications named by the data center manager when resource competition arises according to the time-varying capacity demands and the quality of applications. Using Rainbow, a Xen-based prototype we implemented, we evaluate the VM-based shared platform as well as the two-tiered on-demand resource allocation mechanism and algorithms. The experimental results show that Rainbow without dynamic resource allocation (Rainbow-NDA) provides 26 to 324 percent improvements in the application performance, as well as 26 percent higher average CPU utilization than traditional service computing framework, in which applications use exclusive servers. The two-tiered on-demand resource allocation further improves performance by 9 to 16 percent for those critical applications, 75 percent of the maximum performance improvement, introducing up to 5 percent performance degradations to others, with 1 to 5 percent improvements in the resource utilization in comparison with Rainbow-NDA.

Resource allocation and admission control algorithm based on non-cooperation game in wireless mesh networks