基于混合粒子群算法的计算卸载成本优化

为了满足用户日益增长的计算密集型和时延敏感型服务需求,同时最小化计算任务的处理成本,在时延约束下,该文针对超密集异构边缘计算网络,构建了有关任务卸载、无线资源管理、计算资源块分配的联合优化问题。考虑到所规划的问题具有非线性和混合整数的形式,且为满足约束条件及提升算法收敛速率,通过改进分层自适应搜索(HAS)算法设计了混合粒子群优化 (HPSO)算法来求解所提出的问题。仿真结果表明,HPSO算法明显优于现有算法,能有效降低任务处理成本。     

Cloud-assisted Computing for Event-driven Mobile Services

Today, software developers for desktop computing build request and respond applications to do what end users tell them to do and answer what they ask. In mobile computing, software developers will need to develop sense and response applications that will interact with the end user. These applications will notify or ask users what they want based on what they have sensed or on a personal profile. Mobile cloud computing has the potential to empower mobile users with capabilities not found in mobile devices, combining different and heterogeneous data sets. In this work, we discuss the importance and challenges in designing event-driven mobile services that will detect conditions of interest to users and notify them accordingly.     

边缘算力网络中智能算力感知路由分配策略研究

增强现实、自动驾驶、智慧城市、工业互联网等新型业务应用对网络算力的需求逐渐增强,然而,边缘算力网络系统面临着网络共存的问题——负载不均衡,导致一部分边缘服务器无法满足业务应用的处理需求,另一部分边缘服务器的算力资源处于空闲状态.为了高效协同地感知利用泛在、异构的算力资源,提升6G通信网络的内生感知和算力自适应能力,急需...     

云环境中基于异构资源的资源分配算法

随着"云计算"的出现和快速发展,"云"作为一种新型的资源形式被越来越多的用户所使用。云环境中的资源分配问题成为了云计算中不可忽略的问题。在云资源管理平台中,如何既满足用户的任务需求,又节省云资源成本,是云运营商尽快希望解决的问题之一。实际上云用户对云资源的请求是有差异的,而且用户任务的完成通常由多个异构的云资源来实现。文中作者考虑了异构云资源间的差异,提出了一种基于异构资源的资源分配算法。该算法先从任务的全局角度考虑,将用户提交的云任务划成不同的组合,再根据云资源间的差异,为相应的组合分配相应的资源。实验仿真表明,在异构云环境中,该算法能在满足用户需求的前提下,在节省云资源使用上有较好的表现。     

Adaptive transmitter optimization for blind and group-blind multiuser detection

The linear subspace-based blind and group-blind multiuser detectors recently developed represent a robust and efficient adaptive multiuser detection technique for code-division multiple-access (CDMA) systems. In this paper, we consider adaptive transmitter optimization strategies for CDMA systems operating in fading multipath environments in which these detectors are employed. We make use of more recent results on the analytical performance of these blind and group-blind receivers in the design and analysis of the transmitter optimization techniques. In particular, we develop a maximum-eigenvector-based method of optimizing spreading codes for given channel conditions and a utility-based power control algorithm for CDMA systems with blind or group-blind multiuser detection. We also design a receiver incorporating joint optimization of spreading codes and transmitter power by combining these algorithms in an iterative configuration. We will see that the utility-based power control algorithm allows us to efficiently set performance goals through utility functions for users in heterogeneous traffic environments and that spreading code optimization allows us to achieve these goals with lower transmit power. The signal processing algorithms presented here maintain the blind (or group-blind) nature of the receiver and are distributed, i.e., all power and spreading code adjustments can be made using only locally available information.     

Toward High Utilization of Heterogeneous Computing Resources in SNP Detection

As the amount of re‐sequencing genome data grows, minimizing the execution time of an analysis is required. For this purpose, recent computing systems have been adopting both high‐performance coprocessors and host processors. However, there are few applications that efficiently utilize these heterogeneous computing resources. This problem equally refers to the work of single nucleotide polymorphism (SNP) detection, which is one of the bottlenecks in genome data processing. In this paper, we propose a method for speeding up an SNP detection by enhancing the utilization of heterogeneous computing resources often used in recent high‐performance computing systems. Through the measurement of workload in the detection procedure, we divide the SNP detection into several task groups suitable for each computing resource. These task groups are scheduled using a window overlapping method. As a result, we improved upon the speedup achieved by previous open source applications by a magnitude of 10.     

Practical deniable authentication for pervasive computing environments

Pervasive computing environments allow users to get services anytime and anywhere. Security has become a great challenge in pervasive computing environments because of its heterogeneity, openness, mobility and dynamicity. In this paper, we propose two heterogeneous deniable authentication protocols for pervasive computing environments using bilinear pairings. The first protocol allows a sender in a public key infrastructure (PKI) environment to send a message to a receiver in an identity-based cryptography (IBC) environment. The second protocol allows a sender in the IBC environment to send a message to a receiver in the PKI environment. Our protocols admits formal security proof in the random oracle model under the bilinear Diffie–Hellman assumption. In addition, our protocols support batch verification that can speed up the verification of authenticators. The characteristic makes our protocols useful in pervasive computing environments.     

Multidimensional QoS cloud computing resource scheduling method based on stakeholder perspective

A multidimensional cloud computing architecture is designed and a multidimensional cloud resource scheduling model is constructed based on the stakeholder perspective of cloud users and cloud service providers to meet the high QoS requirements of cloud users (such as task execution time and task completion time) with low computing costs (such as energy consumption,economic costs and system availability).For the second-level cloud resource scheduling,an MQoS cloud resource scheduling algorithm based on multiple Greedy algorithm is proposed.The experimental results show that under the four cloud computing application scenarios with no aftereffects,the MQoS cloud resource scheduling algorithm has an overall increase of 206.42%~228.99% and 34.26%~56.93 in terms of multidimensional QoS degree compared with FIFO and M2EC algorithms.It has an average overall reduction of 0.48~0.49 and 0.20~0.27 in terms of cloud data center load balance difference.     

Run-Time Management of a MPSoC Containing FPGA Fabric Tiles

Multimedia applications, like, e.g., 3-D games and video decoders, are typically composed of communicating tasks. Their target embedded computing platforms (e.g., TI OMAP3, IBM Cell) contain multiple heterogeneous processing elements. At application design-time, it is often unknown which applications will execute simultaneously. Hence, resource assignment decisions need to be made by a run-time manager. Run-time assignment of these communicating tasks onto the communication and computation resources of such a multiprocessor platform is a challenging task. In the presence of fine-grain reconfigurable hardware processing elements, the run-time manager also needs to consider the creation of a so-called configuration hierarchy. Instead of executing a dedicated hardware task, the fine-grain reconfigurable hardware fabric hosts a programmable softcore block that, in turn, executes the task functionality. Hence, the next challenge for run-time management is to efficiently handle a configuration hierarchy. This paper details a run-time task assignment heuristic that performs fast and efficient task assignment in a multiprocessor system-on-chip containing fine-grain reconfigurable hardware tiles. In addition, this algorithm is capable of managing a configuration hierarchy. We show that being capable of handling a configuration hierarchy significantly improves the task assignment performance (i.e., success rate and assignment quality). In several cases, adding a configuration hierarchy improves the assignment success rate of the assignment heuristic by 20%.     

一种基于分层多代理的云计算负载均衡方法

针对现有云计算系统中负载均衡方法的不足,借鉴系统逻辑分层和多代理的思想,提出一种基于分层多代理的云计算负载均衡方法.通过对云计算平台逻辑分层,在任务代理层设置任务监控代理和任务子代理,根据用户任务的差异性,采用基于任务优先级和QoS目标约束的调度策略协同完成任务调度;在资源代理层设置资源监控代理和资源子代理,考虑物理节点的异构性,采用基于启发式贪婪的资源分配策略协同完成虚拟机到物理节点的映射.通过评估对比仿真实验,结果表明该方法在任务调度效率、任务完成时间、截止时间违背率和负载均衡度方面表现更优,多代理有效地分担了中心管理节点的管理负载,使云计算平台的任务处理能力、资源利用率及鲁棒性均得到了进一步的提升.     

移动边缘计算中分布式异构任务卸载算法

随着物联网(IoT)迅速发展,移动边缘计算(MEC)在提供高性能、低延迟计算服务方面的作用日益明显。然而,在面向IoT业务的MEC(MEC-IoT)时变环境中,不同边缘设备和应用业务在时延和能耗等方面具有显著的异构性,对高效的任务卸载及资源分配构成严峻挑战。针对上述问题,该文提出一种动态的分布式异构任务卸载算法(D2HM),该算法利用分布式博弈机制并结合李雅普诺夫优化理论,设计了一种资源的动态报价机制,并实现了对不同业务类型差异化控制和计算资源的弹性按需分配,仿真结果表明,所提的算法可以满足异构任务的多样化计算需求,并在保证网络稳定性的前提下降低系统的平均时延。     

A blockchain-enabled security management framework for mobile edge computing

Mobile edge computing (MEC) integrates mobile and edge computing technologies to provide efficient computing services with low latency. It includes several Internet of Things (IoT) and edge devices that process the user data at the network's edge. The architectural characteristic of MEC supports many internet-based services, which attract more number of users, including attackers. The safety and privacy of the MEC environment, especially user information is a significant concern. A lightweight accessing and sharing protocol is required because edge devices are resource constraints. This paper addresses this issue by proposing a blockchain-enabled security management framework for MEC environments. This approach provides another level of security and includes blockchain security features like temper resistance, immutable, transparent, traceable, and distributed ledger in the MEC environment. The framework guarantees secure data storage in the MEC environment. The contributions of this paper are twofold: (1) We propose a blockchain-enabled security management framework for MEC environments that address the security and privacy concerns, and (2) we demonstrate through simulations that the framework has high performance and is suitable for resource-constrained MEC devices. In addition, a smart contract-based access and sharing mechanism is proposed. Our research uses a combination of theoretical analysis and simulation experiments to demonstrate that the proposed framework offers high security, low latency, legitimate access, high throughput, and low operations cost.     

基于二次规划的任务卸载决策和资源分配方法

主要研究移动用户均有多个独立任务的多用户移动云计算系统,这些移动用户将任务卸载到云端时共享通信资源。如何对所有用户的任务卸载决策和通信资源分配进行联合优化,以便使所有用户的能耗、计算量和延时降到最低是目前研究的难点。将该问题建模为NP难度的非凸的具有二次约束的二次规划(QCQP)问题,提出一种高效的近似算法进行求解,通过单独的半正定松驰(SDR)处理后,确定二元卸载决策和通信资源最优分配。采用代表最小系统成本的性能下界作为性能基准进行仿真实验,结果表明,本文算法在多种参数配置下的性能均接近最优性能。     

Task-oriented resource prediction and adjustment algorithm for space-based information network

The tasks of a space-based information network are complex and diverse, but the resources of a space-based environment are minimal. The existing methods are challenging to match task demand to resource supply accurately. Aiming at the problem of accurate prediction from task to resource, we propose a resource prediction adjustment strategy. First, we propose a multidimensional resource prediction algorithm based on improved particle swarm optimization and back propagation (IPSO-BP) neural network. The improved PSO is used to optimize the weight and threshold of BP neural network to make up for the defects that BP neural network is easy to fall into local minimum and the predicted output value is not unique. Second, to meet the quality of service (QoS) of tasks, we propose a density-based performance evaluation algorithm (DPEA) to adjust resources. This method uses the idea of local sensitive hash to select the evaluation subset for the configuration task, then dynamically selects the $k$ nearest neighbors of the configuration task, and uses the idea of weighted average to evaluate the QoS performance index of the configuration task. Simulation results show that the proposed resource prediction and adjustment strategy effectively reduces the scheduling time overhead and improves resource utilization.     

A genetic algorithm‐based task scheduling for cloud resource crowd‐funding model

With the rapid development of cloud computing, the number of cloud users is growing exponentially. Data centers have come under great pressure, and the problem of power consumption has become increasingly prominent. However, many idle resources that are geographically distributed in the network can be used as resource providers for cloud tasks. These distributed resources may not be able to support the resource‐intensive applications alone because of their limited capacity; however, the capacity will be considerably increased if they can cooperate with each other and share resources. Therefore, in this paper, a new resource‐providing model called “crowd‐funding” is proposed. In the crowd‐funding model, idle resources can be collected to form a virtual resource pool for providing cloud services. Based on this model, a new task scheduling algorithm is proposed, RC‐GA (genetic algorithm for task scheduling based on a resource crowd‐funding model). For crowd‐funding, the resources come from different heterogeneous devices, so the resource stability should be considered different. The scheduling targets of the RC‐GA are designed to increase the stability of task execution and reduce power consumption at the same time. In addition, to reduce random errors in the evolution process, the roulette wheel selection operator of the genetic algorithm is improved. The experiment shows that the RC‐GA can achieve good results.     

Enforcing Cooperative Resource Sharing in Untrusted P2P Computing Environments

Peer-to-Peer (P2P) computing is widely recognized as a promising paradigm for building next generation distributed applications. However, the autonomous, heterogeneous, and decentralized nature of participating peers introduces the following challenge for resource sharing: how to make peers profitable in the untrusted P2P environment? To address the problem, we present a self-policing and distributed approach by combining two models: PET, a personalized trust model, and M-CUBE, a multiple-currency based economic model, to lay a foundation for resource sharing in untrusted P2P computing environments. PET is a flexible trust model that can adapt to different requirements, and provides the solid support for the currency management in M-CUBE. M-CUBE provides a novel self-policing and quality-aware framework for the sharing of multiple resources, including both homogeneous and heterogeneous resources. We evaluate the efficacy and performance of this approach in the context of a real application, a peer-to-peer Web server sharing. Our results show that our approach is flexible enough to adapt to different situations and effective to make the system profitable, especially for systems with large scale.     

Efficient task scheduling in cloud networks using ANN for green computing

Recently, there has been a growing emphasis on reducingenergy consumption in cloud networks and achieving green computing practices toaddress environmental concerns and optimize resource utilization. In thiscontext, efficient task scheduling minimizes energy usage and enhances overallsystem performance. To tackle the challenge ofenergy-efficient task allocation, we propose a novel approach that harnessesthe power of Artificial Neural Networks (ANN). Our Artificial neural network Dynamic Balancing (ANNDB) method is designed toachieve green computing in cloud environments. ANNDB leverages the feed-forwardnetwork architecture and a multi-layer perceptron, effectively allocatingrequests to higher-power and higher-quality virtual machines, resulting inoptimized energy utilization. Through extensive simulations, wedemonstrate the superiority of ANNDB over existing methods, including WPEG,IRMBBC, and BEMEC, in terms of energy and power efficiency. Specifically, ourproposed ANNDB method exhibits substantial improvements of 13.81%, 8.62%, and9.74% in the Energy criterion compared to WPEG, IRMBBC, and BEMEC,respectively. Additionally, in the Power criterion, the method achievesperformance enhancements of 3.93%, 4.84%, and 4.19% over the mentioned methods.The findings from this research hold significant promise for organizations seekingto optimize their cloud computing environments while reducing energyconsumption and promoting sustainable computing practices. By adopting theANNDB approach for efficient task scheduling, businesses and institutions cancontribute to green computing efforts, reduce operational costs, and make moreenvironmentally friendly choices without compromising task allocationperformance.     

基于NOMA的超密集MEC网络任务卸载和资源分配方案

为了提高移动边缘计算(mobile edge computing,MEC)网络的频谱效率,满足大量用户的服务需求,建立了基于非正交多址接入(non-orthogonal multiple access,NOMA)的超密集MEC系统模型。为了解决多个用户同时卸载带来的严重通信干扰等问题,以高效利用边缘服务器资源,提出了一种联合任务卸载和资源分配的优化方案,在满足用户服务质量的前提下最小化系统总能耗。该方案联合考虑了卸载决策、功率控制、计算资源和子信道资源分配。仿真结果表明,与其他卸载方案相比,所提方案可以在满足用户服务质量的前提下有效降低系统能耗。     

基于深度学习的异构资源分配算法研究

资源分配是目前云计算领域中一个重要的研究方向。在异构云计算体系结构下的复杂应用问题研究中,为了满足异构资源分配的需求,提升资源利用效率,文中提出了一种基于深度学习的面向应用的资源分配算法。该算法将数据特征进行量化,更加精确地刻画了不同服务器资源之间的性能差异,在分配算法中加入了一个工作负载预测模型,使给出的资源分配方案与需求更加匹配,同时提高了资源利用率。