基于服务器实时性能状态的动态负载均衡算法

为实现Web服务器集群合理的作业任务分配,文章提出了一种新的负载均衡算法,综合考虑了负载均衡调度器后端的业务主机的实时性能,实现了负载均衡调度的动态调整.     

基于流媒体服务器集群副本放置策略的研究

网络流媒体服务具有数据流量大、服务时间长的特点,同时,高并发的服务请求会造成流媒体服务质量明显下降等性能瓶颈问题。针对这种现象,设计了一种基于Zipf分布规律的两阶段副本放置策略,并成功应用于某市中小学心理辅导的视讯平台。实践结果表明,与传统的方法相比,该方法具有较好的实用值和健壮性。     

Scalable network file systems with load balancing and fault tolerance for web services

Because of the rapid growth of the World Wide Web and the popularization of smart phones, tablets and personal computers, the number of web service users is increasing rapidly. As a result, large web services require additional disk space, and the required disk space increases with the number of web service users. Therefore, it is important to design and implement a powerful network file system for large web service providers. In this paper, we present three design issues for scalable network file systems. We use a variable number of objects within a bucket to decrease internal fragmentation in small files. We also propose a free space and access load-balancing mechanism to balance overall loading on the bucket servers. Finally, we propose a mechanism for caching frequently accessed data to lower the total disk I/O. These proposed mechanisms can effectively improve scalable network file system performance for large web services.     

Adaptive and scalable load balancing for metadata server cluster in cloud-scale file systems

Big data is an emerging term in the storage industry, and it is data analytics on big storage, i.e., Cloud-scale storage. In Cloud-scale (or EB-scale) file systems, load balancing in request workloads across a metadata server cluster is critical for avoiding performance bottlenecks and improving quality of services.Many good approaches have been proposed for load balancing in distributed file systems. Some of them pay attention to global namespace balancing, making metadata distribution across metadata servers as uniform as possible. However, they do not work well in skew request distributions, which impair load balancing but simultaneously increase the effectiveness of caching and replication. In this paper, we propose Cloud Cache (C²), an adaptive and scalable load balancing scheme for metadata server cluster in EB-scale file systems. It combines adaptive cache diffusion and replication scheme to cope with the request load balancing problem, and it can be integrated into existing distributed metadata management approaches to efficiently improve their load balancing performance. C² runs as follows: 1) to run adaptive cache diffusion first, if a node is overloaded, loadshedding will be used; otherwise, load-stealing will be used; and 2) to run adaptive replication scheme second, if there is a very popular metadata item (or at least two items) causing a node be overloaded, adaptive replication scheme will be used, in which the very popular item is not split into several nodes using adaptive cache diffusion because of its knapsack property. By conducting performance evaluation in trace-driven simulations, experimental results demonstrate the efficiency and scalability of C².     

针对固定码率(CBR)流媒体的动态替换算法(DRA)

流媒体代理服务器缓存是针对流媒体访问的一项技术,能有效的提高流媒体访问质量。缓存算法是缓存代理服务器的重要技术组成部分,包括缓存保存方式、替换算法及进入策略三个方面。文中将介绍一套针对固定码率(CBR)流媒体的缓存算法,在包含码率分级保存方法的同时,使用了包含流媒体缓存基本特性的动态替换算法(DRA,Dynamic Replication Algorithm)。     

Parallel extremal optimization in processor load balancing for distributed applications

The paper concerns parallel methods for extremal optimization (EO) applied in processor load balancing in execution of distributed programs. In these methods EO algorithms detect an optimized strategy of tasks migration leading to reduction of program execution time. We use an improved EO algorithm with guided state changes (EO-GS) that provides parallel search for next solution state during solution improvement based on some knowledge of the problem. The search is based on two-step stochastic selection using two fitness functions which account for computation and communication assessment of migration targets. Based on the improved EO-GS approach we propose and evaluate several versions of the parallelization methods of EO algorithms in the context of processor load balancing. Some of them use the crossover operation known in genetic algorithms. The quality of the proposed algorithms is evaluated by experiments with simulated load balancing in execution of distributed programs represented as macro data flow graphs. Load balancing based on so parallelized improved EO provides better convergence of the algorithm, smaller number of task migrations to be done and reduced execution time of applications.     

一种适用于无线网络的流媒体传输机制

为保证无线网络中多媒体数据的传输质量,提出了一种适用于无线网络的流媒体传输机制（WMTCC）。该机制通过发送探测报文区分网络拥塞丢包和链路误码随机丢包,准确判断网络的拥塞状况,实施发送速率调节,保证了流媒体服务质量(QoS)。由于准确区分出无线链路误码丢包,该机制在链路误码率较高时能维持较高的网络吞吐量。仿真实验结果显示在高误码率无线网络中,该机制可以获得更高的吞吐量和更大的拥塞窗口,并且发送速率的变化更加平滑。     

Asymmetry-aware load balancing for parallel applications in single-ISA multi-core systems

Contemporary operating systems for single-ISA (instruction set architecture) multi-core systems attempt to distribute tasks equally among all the CPUs. This approach works relatively well when there is no difference in CPU capability. However, there are cases in which CPU capability differs from one another. For instance, static capability asymmetry results from the advent of new asymmetric hardware, and dynamic capability asymmetry comes from the operating system (OS) outside noise caused from networking or I/O handling. These asymmetries can make it hard for the OS scheduler to evenly distribute the tasks, resulting in less efficient load balancing. In this paper, we propose a user-level load balancer for parallel applications, called the ’capability balancer’, which recognizes the difference of CPU capability and makes subtasks share the entire CPU capability fairly. The balancer can coexist with the existing kernel-level load balancer without detrimenting the behavior of the kernel balancer. The capability balancer can fairly distribute CPU capability to tasks with very little overhead. For real workloads like the NAS Parallel Benchmark (NPB), we have accomplished speedups of up to 9.8% and 8.5% in dynamic and static asymmetries, respectively. We have also experienced speedups of 13.3% for dynamic asymmetry and 24.1% for static asymmetry in a competitive environment. The impacts of our task selection policies, FIFO (first in, first out) and cache, were compared. The use of the cache policy led to a speedup of 5.3% in overall execution time and a decrease of 4.7% in the overall cache miss count, compared with the FIFO policy, which is used by default.     

Numerical studies on a paradox for non-cooperative static load balancing in distributed computer systems

Numerical examples of a Braess-like paradox in which adding capacity to a distributed computer system may degrade the performance of all users in the system under non-cooperative optimization have been reported. Unlike the original Braess paradox, in the models examined, this behavior occurs only in the case of finitely many users and not in the case of infinite number of users and the degree of performance degradation can increase without bound. This study examines numerically some examples around the Braess-like paradox in a distributed computer system. In the numerical examples, it is observed that the worst-case degree of the paradox (WCDP) is largest in complete symmetry. The dependence of the WCDP on some system parameters is also examined.     

面向TCP友好性的IP网络实时流媒体流预测控制

为了解决IP网络实时流媒体流控中既要避免网络拥塞又要将接收端缓冲区长度控制在合理范围的难题, 本文提出了一种面向TCP友好性的IP网络实时流媒体流预测控制算法. 该算法将IP网络实时流媒体流控转化为具有TFRC约束时变不确定条件的预测控制问题, 使用改进的动态矩阵控制算法(DMC)进行优化求解. 仿真实验结果表明该方法能够适应IP网络的复杂环境, 满足了TCP友好性要求, 有效地消除了网络传输抖动对播放带来的不利影响, 具有良好的实用性.     

An ant colony optimization algorithm for load balancing in parallel machines with sequence-dependent setup times

This study introduces the problem of minimizing average relative percentage of imbalance (ARPI) with sequence-dependent setup times in a parallel-machine environment. A mathematical model that minimizes ARPI is proposed. Some heuristics, and two metaheuristics, an ant colony optimization algorithm and a genetic algorithm are developed and tested on various random data. The proposed ant colony optimization method outperforms heuristics and genetic algorithm. On the other hand, heuristics using the cumulative processing time obtain better results than heuristics using setup avoidance and a hybrid rule in assignment.     

Proximal policy optimization-based controller for chaotic systems

Deep reinforcement learning (DRL) algorithms are suitable for modeling and controlling complex systems. Methods for controlling chaos, a difficult task, require improvement. In this article, we present a DRL-based control method that can control a nonlinear chaotic system without any prior knowledge of the system's equations. We use proximal policy optimization (PPO) to train an agent. The environment is a Lorenz chaotic system, and our goal is to stabilize this chaotic system as quickly as possible and minimize the error by adding extra control terms to the chaotic system. Therefore, the reward function accounts for the total triaxial error. The experimental results demonstrated that the trained agent can rapidly suppress chaos in the system, regardless of the system's random initial conditions. A comprehensive comparison of different DRL algorithms indicated that PPO is the most efficient and effective algorithm for controlling the chaotic system. Moreover, different maximum control forces were applied to determine the relationship between the control forces and controller performance. To verify the robustness of the controller, random disturbances were introduced during training and testing, and the empirical results indicated that the agent trained with random noise performed better. The chaotic system has highly nonlinear characteristics and is extremely sensitive to initial conditions, and DRL is suitable for modeling such systems.     

动态环境下分布式异构多机器人避障方法研究

多机器人系统在联合搜救、智慧车间、智能交通等领域得到了日益广泛的应用。目前,多个机器人之间、机器人与动态环境之间的路径规划和导航避障仍需依赖精确的环境地图,给多机器人系统在非结构环境下的协调与协作带来了挑战。针对上述问题,本文提出了不依赖精确地图的分布式异构多机器人导航避障方法,建立了基于深度强化学习的多特征策略梯度优化算法,并考虑了人机协同环境下的社会范式,使分布式机器人能够通过与环境的试错交互,学习最优的导航避障策略;并在Gazebo仿真环境下进行了最优策略的训练学习,同时将模型移植到多个异构实体机器人上,将机器人控制信号解码,进行真实环境测试。实验结果表明:本文提出的多特征策略梯度优化算法能够通过自学习获得最优的导航避障策略,为分布式异构多机器人在动态环境下的应用提供了一种技术参考。