A Fuzzy Logic-Based Method for Replica Placement in the Peer to Peer Cloud Using an Optimization Algorithm

The Peer to Peer-Cloud (P2P-Cloud) is a suitable alternative to distributed cloud-based or peer-to-peer (P2P)-based content on a large scale. The P2P-Cloud is used in many applications such as IPTV, Video-On-Demand, and so on. In the P2P-Cloud network, overload is a common problem during overcrowds. If a node receives many requests simultaneously, the node may not be able to respond quickly to user requests, and this access latency in P2P-Cloud networks is a major problem for their users. The replication method in P2P-Cloud environments reduces the time to access and uses network bandwidth by making multiple data copies in diverse locations. The replication improves access to the information and increases the reliability of the system. The data replication's main problem is identifying the best possible placement of replica data nodes based on user requests for data access time and an NP-hard optimization problem. This paper proposes a new replica replacement to improve average access time and replica cost using fuzzy logic and Ant Colony Optimization algorithm. Ants can find the shortest path to discover the optimal node to place the duplicate file with the least access time latency. The fuzzy module evaluates the historical information of each node to analyze the pheromone value per iteration. The fuzzy membership function is also used to determine each node's degree based on the four characteristics. The simulation results showed that the access time and replica cost are improved compared to other replica replacement algorithms.

     

Performance Improvement of MapReduce for Heterogeneous Clusters Based on Efficient Locality and Replica Aware Scheduling (ELRAS) Strategy

MapReduce is a parallel programming model for processing the data-intensive applications in a cloud environment. The scheduler greatly influences the performance of MapReduce model while utilized in heterogeneous cluster environment. The dynamic nature of cluster environment and computing workloads affect the execution time and computational resource usage in the scheduling process. Further, data locality is essential for reducing total job execution time, cross-rack communication, and to improve the throughput. In the present work, a scheduling strategy named efficient locality and replica aware scheduling (ELRAS) integrated with an autonomous replication scheme (ARS) is proposed to enhance the data locality and performs consistently in the heterogeneous environment. ARS autonomously decides the data object to be replicated by considering its popularity and removes the replica as it is idle. The proposed approach is validated in a heterogeneous cluster environment with various realistic applications that are IO bound, CPU bound and mixed workloads. ELRAS improves the throughput by a factor about 2 as compared with the existing FIFO and it also yields near optimal data locality, reduce the execution time, and effective utilization of resources. The simplicity of ELRAS algorithm proves its feasibility to adopt for a wide range of applications.

     

Method for replica creation in data grids based on complex networks

In order to reduce makespan and storage consumption in data grids,a node selection model for replica creation is proposed.The model is based on the degree distribution of complex networks.We define two candidate replica nodes: a degree-based candidate pool and a frequency-based candidate pool,through which a degree-based candidate pool is defined m consideration of onsidering the access frequency,a candidate pool-based frequency is also defined.The data replica is copied to the node with the minimum local cost in the two pools.Further,this paper presents and proves a replica creation theorem.A dynamic multi-replicas creation algorithm(DMRC)is also provided.Simulation results show that the proposed method may simultaneously reduce makespan and data used in space storage consumption.     

基于数据中心负载分析的自适应延迟调度算法

由于已有的延迟调度算法基于静态的等待时间阈值,因此不能适应云计算数据中心动态的负载变化。针对该问题,提出了等待时间阈值自适应调整模型。基于该模型,设计了自适应延迟调度算法(ADS)。ADS算法通过分析空闲计算节点的到达强度、网络带宽和作业执行状态等参数,自适应调整等待时间阈值,以减少作业响应时间。基于开发的原型系统,验证了自适应调整模型,测试了算法性能。结果表明,ADS算法在作业响应时间等方面优于已有的延迟调度算法。     

QoS-aware replica placement for data intensive applications

In recent years, with the rapid development of data intensive applications, data replication has become an enabling technology for the data grid to improve data availability, and reduce file transfer time and bandwidth consumption. The placement of replicas has been proven to be the most difficult problem that must be solved to realize the process of data replication. This paper addresses the quality of service （QoS） aware replica placement problem in data grid, and proposes a dynamic programming based replica placement algorithm that not only has a QoS requirement guarantee, but also can minimize the overall replication cost, including storage cost and communication cost. By simulation, experiments show that the replica placement algorithm outperforms an existing popular replica placement technique in data grid.     

Stochastic job stream scheduling method for cipher chip with multi-cryptography

Aiming at the rich of safety requirements of tasks which resulting in random cross access to multi cipher algorithms, a hierarchical hardware scheduling method was presented with associated control based on data identifica-tion. The first level was responsible for distributing tasks to different cipher clusters, and by optimizing the search logic to achieve rapid distribution of data. The second level was responsible for completing the context-related tasks in scheduling order by adding an association control module and association queues. Intermediate state storage module realized the saving of the intermediate state in serial cipher algorithm modes, which was indexed by task ID. Pre-processing module process data inputted by the succeeding tasks. It is proved that the proposed scheduling algo-rithm solves the problem of random cross encryption and decryption in many-to-many communication model of high-speed data stream.     

面向众核密码处理器的高效负载均衡技术

工作负载分配不均是制约众核密码平台资源利用率提高的重要因素,动态负载分配可提高平台资源利用率,但具有一定开销;所以更高的负载均衡频率并不一定带来更高的负载均衡增益。因此,该文建立了关于负载均衡增益率与负载均衡频率的数学模型。基于模型,提出一种面向众核密码平台的无冲突负载均衡策略和一种基于硬件作业队列的“可扩展-可移植”负载均衡引擎——“簇间微网络-簇内环阵列”。实验证明:在性能、延时功耗积、资源利用率和负载均衡度方面,该文设计的负载均衡引擎与基于“作业窃取”的软件技术相比平均优化约4.06倍、7.17倍、23.01%和2.15倍;与基于“作业窃取”的硬件技术相比约优化1.75倍、2.45倍、10.2%、和1.41倍;与理想硬件技术相比,密码算法吞吐率平均只降低了约5.67%(最低3%)。实验结果表明该文技术具有良好的可扩展性和可移植性。     

MR-MC无线传感器网络最小延迟数据聚集调度研究

单radio单信道无线传感器网络的最小延迟聚集调度是一个NPC问题,已提出许多解决方案。在多radio多信道网络中,节点可以同时接收多个不同节点传输的数据,降低延迟。基于上述特点,考虑树结构约束,时槽、信道和radio分配等约束条件,将多radio多信道无线传感器网络最小延迟聚集调度问题定义为一个优化问题,并分解为建立聚集树和节点调度2个子问题,针对这2个子问题分别提出启发式算法。实验结果表明,提出的算法具有良好的性能。     

Data Replication in Cooperative Mobile Ad-Hoc Networks

The mobile computing environment provides many benefits such as ubiquitous access to computing but includes constraints on resources such as available bandwidth and battery life. Replication is a widely recognized method for balancing the demands of storage space with bandwidth and battery life. We propose a novel scheme that seeks to strategically balance these constrained resources through a cooperative game-theory approach for replication in a mobile environment. Our replication strategy relies on the cooperation of the nodes within the network to make replica caching decisions which are spatiotemporally local-optimal for the network from an energy and bandwidth conservation standpoint. In cooperative altruistic data replication, each node calculates the net global benefit, for caching a replica of the requested data, as the result data is returned from the responding node to the requesting node, where it determines the spatiotemporally local-optimal node for replicating the data item. Performance results from our research indicate that our scheme, CADR, improves the query response time by 25 and 45 %, mean hop count is improved by 26 and 46 %, query error is reduced by 30 and 48 %, while energy utilization is reduced 30 and 57 % when compared with both another game theoretic replication approach and standard cooperative caching respectively.     

Placement problems for transparent data replication proxy services

Transparent data replication has been considered a promising technique for improving system performance for a large distributed network. In this paper, a hybrid transparent replication model is presented. We address the problems of replication proxy placement in the network and data replica placement on the installed proxies given that a maximum of M proxies are allowed. Both reads and writes are considered in these problems. The performance objective is to minimize the total data transfer cost. To address the placement problems, we first present the optimal solutions for a single object in a tree network without/with constraint on the number of replicas. Based on that, two schemes, namely, aggregate access (AGGA) and weighted popularity (WPOP), are proposed for the replication proxy placement problem. An optimal solution is described for the replica placement problem. The performance of the proposed placement schemes is evaluated with a set of carefully designed simulation experiments over a wide range of system parameters. The results give us several helpful intuitions in deploying transparent replication proxies in a practical system.     

云环境下面向跨域作业的调度方法

云环境下,因数据局部性或是任务对资源的特殊偏好,一个作业所包含的任务往往需要在不同的数据中心局点上运行,此类作业称为跨域作业.跨域作业的完成时间取决于最慢任务的执行效率,即存在木桶效应.针对各域资源能力异构条件下不合理的调度策略导致跨域作业执行时间跨度过长的问题,本文提出一种面向跨域作业的启发式调度方法MIN-Max-Min,优先选择期望完成时间最短的作业执行.通过实验表明,与先来先服务的策略相比,该方法能将跨域作业平均执行时间跨度减少40%以上.     

基于粒子群优化算法的数据中心网络流量调度策略

为解决当前数据中心网络存在链路负载不均衡及带宽资源浪费问题,提出了一种基于粒子群优化算法的流量调度策略.该策略结合软件定义网络控制器可获取全局网络拓扑信息的特性,依据当前链路带宽资源状况及网络流量的带宽需求建立目标函数.首先,根据流的源地址和目的地址找出最短路径集,通过定义粒子聚合度判断算法是否有陷入局部最优的趋势;然后,结合约束条件与目标函数,利用优化的粒子群算法从最短路径集中找出网络流量的最佳调度路径.实验结果表明,相比于其他算法,该算法有效地提高了网络平均吞吐量,获取了较低的丢包率,从而减轻了带宽资源的浪费,更好地实现了网络的负载均衡.     

具负载平衡之IPTV影音串流系统

影音媒体具高带宽及实时特性,因此影音串流系统的服务质量的提升端赖影音服务器的有效管理。在此提出一套Video replication及Access request distribution的算法,藉由将热门影音分散到各影音服务器以及影音存取的分散,以达到各影音服务器系统负载平衡的效果。在此比较了3个不同的算法的负载平衡的状况,经实例验证,提出的算法确能达到影音服务器负载平衡的效果,承载更多的系统服务要求。该系统构建在Smart-i TV的IPTV系统台,并讨论了实作的相关技术。     

基于节点识别的慢任务调度算法

为了降低大数据处理集群在执行任务过程中的慢任务对作业执行效率的影响,提出了一种识别慢任务、备份慢任务、减少慢任务相结合的调度算法——TQST算法。首先,通过判断节点能力和任务执行时间,建立慢节点、非常慢节点和慢任务队列;其次,根据预判备份执行价值确定如何启动慢任务的备份任务,提高了备份执行的作用;然后,在节点识别的基础上,规避为非常慢节点分配任务,从根本上减少慢任务的产生,提高作业执行效率。实验结果表明,TQST算法在作业响应时间等方面优于已有的慢任务调度算法。     

无中心式调度框架下网格作业的节点选择优化

为优化无中心式调度框架下网格作业的节点选择,提出了随机多起点爬山算法.为使多个起点均匀分布于网格,按随机选择邻居的重复次数的指数增长找出各起点.为反映合理的用户调度需求,用平均的并行计算能力加权的有界减慢率衡量节点选择.灵活调整网格工作负荷,对随机多起点爬山算法进行了全面评估.在网格负载不是很轻情况下,该算法能有效地在网格全局优化节点选择.     

Effects of cache mechanism on wireless data access

In wireless data transmission, the capacity of wireless links is typically limited. Since many applications exhibit temporal locality for data access, the cache mechanism can be built in a wireless terminal to effectively reduce the data access time. This paper studies the cache performance of the wireless terminal by considering a business-card application. We investigate the least-recently used replacement policy and two strongly consistent data access algorithms called poll-each-read and callback. An analytic model is proposed to derive the effective hit ratio of data access, which is used to validate against simulation experiments. Our study reports how the data access rate and the data update distribution affect the cache performance in a wireless terminal.     

云存储系统中数据副本服务的可靠性保障研究

以数据节点与网络链路的可靠性因素分析为基础,提出了云存储系统的数据副本服务可靠性模型。根据访问可靠性与数据副本数量、用户访问量之间的关系,设计数据服务可靠性、副本生成时机、存储节点选择的确定方法,实现了副本分布、删除算法,并在云存储系统ERS-Cloud上进行一系列实验,结果表明该方法能够有效保障数据服务的可靠性,进一步降低副本的冗余存储数量。     

一种基于访问频率的非结构化P2P复制策略

旨在通过数据复制来提高非结构化P2P系统的数据访问效能,提出了基于访问频率的复制策略—AFRE算法,该算法通过赋予不同时间片的访问记录不同的权重,计算数据的局部访问频率,并将适量的数据副本放置到节点度较大的节点.实验证明,提出的复制策略在副本总数明显减少的情况下,能够获得较好的平均查询长度及优于其他常用复制策略的查询成功率.     

Local Coordination Based Routing and Spectrum Assignment in Multi-hop Cognitive Radio Networks

Although Cognitive Radio technology brings efficient spectrum usage and effective interference avoidance, it also brings new challenges to routing in multi-hop Cognitive Radio Networks. Firstly, spectrum assignment is required for each hop in routing; secondly, new delay is introduced during multi-frequency scheduling and frequency switching in each node; thirdly, the intersecting nodes serving multi-frequency traffic is easy to be bottleneck in neighborhood region. In this paper, we analysis and model the per-node delay and the path delay in multi-hop Cognitive Radio Network. Then we propose a framework of local coordination based routing and spectrum assignment to solve above problems, which consists of one protocol for routing path and one scheme for neighborhood region. A on-demand Routing and Spectrum Assignment Protocol is proposed to exchange the local spectrum information and interact with multi-frequency scheduling in each node. A local coordination scheme is presented to support flow redirection at an intersecting node and distribute heavy multi-frequency workload to its neighborhood. We prove the correctness and effectiveness of the protocol by thorough simulations, and find that the proposed solution provides good adaptability to varying spectrum distribution. The end-to-end delay when adaptive relay is cooperating with routing protocol outperforms traditional bare-routing solutions.     

Using optorsim to efficiently simulate replica placement strategies

Optorsim provides simulations of file replication strategies such as replica placement,replication scheduling,replica consistency maintenance,etc.However,to assess a replica strategy a researcher must write network configuration files and repeatedly modify parameters,which is inefficient.In this article,a scale-free algorithm is developed to generate network topology which is loaded into Optorsim,and a graphical user interface(GUI) is implemented to set all configuration parameters and algorithm parameters....