基于时间序列分析的杀手级任务在线识别方法

通过分析Google集群中任务的失效次数和失效模式,找到具有高失效频次和连续失效特征的杀手级任务。杀手级任务不仅影响云计算系统上应用运行的可靠性与可用性,而且会浪费大量资源并显著增加调度负载。在杀手级任务资源使用模式的基础上,提出一种基于时间序列的在线识别方法,以利用资源使用时间序列在失效早期准确识别出杀手级任务并通知云计算系统采取前摄性失效恢复措施,从而避免不必要的重复调度和资源浪费。实验结果表明,该方法能够以98.5%的准确率在平均3%的失效时间内识别出杀手级任务,同时节约96.75%的系统资源。     

Effects of age,thumb length and screen size on thumb movement coverage on smartphone touchscreens

This study investigated the relationship between the movement coverage on smartphone touchscreens and the factors (age, thumb length and screen size) affecting this. By referring to the thumb movement in the adduction-abduction orientation with a right-hand phone-holding posture, the thumb-coverage area that represents how far the thumb can reach and the centre of gravity in this coverage area were determined. The assessment focused on a comparison of these indices between ages, thumb lengths and screen sizes. The results showed that elderly users and those with longer thumbs are likely to leave more unreachable space at the right side and bottom of touchscreens. Moreover, the thumb-coverage area actually increased when the touchscreen size was increased; however, increasing the size of a smartphone touchscreen does not necessarily increase the thumb-coverage area at the same ratio as the touchscreen size increases.     

基于Hadoop视频转码缓存策略的研究

针对实时云转码系统中已转码的媒体文件重新使用率不高的情况,引入转码缓存机制,提出一种缓存替换策略。采用云存储空间缓存视频,根据视频资源的流行度、请求间隔、视频大小和未来被访问概率计算视频的效能函数值,当缓存空间不足时替换出效能值较小的视频。实验结果表明,与传统置换算法LRU、LFU和FIFO相比,该策略的资源请求命中率分别提高了2?6％,4?3％和9?2％,在一定程度上缩短了视频分发延迟,减少了用户响应时间。     

多层次的Android系统权限控制方法

随着Android智能平台的普及,其安全问题日益受到人们关注.在底层安全方面,部分root工具已经实现了对最新版本Android的root提权,从而给恶意软件滥用权限造成可乘之机;在上层应用安全方面,目前还没有能够在应用权限进行有效管理的方法.基于安全策略的思想,提出了一种Android应用权限动态管理机制,利用安全策略对授权进行描述,在Android框架层设置权限检查点,并调用请求评估算法进行授权评估,从而实现对应用行为的监控.实验结果表明,该方法能够有效管理Android应用权限的正常调用,约束非法调用,并且系统开销较小.     

离散事件系统部分可诊断性分析

针对离散事件系统部分可诊断性问题,提出一种量化评价与分析方法。该方法以树状结构的故障模型为基础,引入可诊断度与可诊断深度指标,能够从可诊断故障覆盖程度与精确程度两个方面对系统可诊断性进行评价,其优点是评价结果量化表示,能为部分可诊断系统的进一步评价、分析与对比提供参考。此外,还讨论了故障模型对系统可诊断度与可诊断深度两个评价指标的影响,并给出了故障模型构造的一般原则。实例分析与讨论结果表明,所提出的可诊断度与可诊断深度指标能够准确反映系统在特定故障模型下的部分可诊断状态。所提出的部分可诊断性评价方法能为基于离散事件模型的复杂系统设计与评价提供依据,并能够进一步为智能、自适应和自愈系统的设计提供参考。     

Flexible design and implementation of QC-Based LDPC decoder architecture for on-line user-defined matrix downloading and efficient decoding

This paper proposes a flexible QC-LDPC decoder to support on-line matrix downloading. Its applications include IEEE 802.16e, 802.11n/ac/ax, and other advanced standards. Via TSMC 40-nm technology, the flexible architecture only has an area of 0.415 mm². The according power dissipation is 84.68 mW at maximal clock frequency of 1.0 GHz. It is successfully verified with 3 LDPC codes pre-defined in IEEE 802.11n/ac/ax.     

一种改进型FBT时钟树结构

针对混合型鱼骨平衡树(FBT)时钟结构的优缺点,结合宏单元的特性,提出了一种针对触发器与宏单元共存的改进型FBT时钟树结构,并总结出一种快速实现该时钟树的方法。在相同条件下,采用该方法实现的时钟树结构与二叉树型、鱼骨型时钟结构和传统的FBT时钟树结构进行比较,结果显示:鱼骨型时钟结构的时序质量最差;改进型FBT时钟树比二叉树型时钟树减少了15%的时钟延时和35%的时钟偏差,且整个过程的实现时间是传统FBT时钟树的30%。     

基于LSTM的S7协议模糊测试用例生成方法

基于传统模糊测试框架的S7协议模糊测试技术存在构造困难和代码覆盖率低的问题,对测试效率和质量产生很大影响。借助神经网络模型对数据较强的学习能力和预测能力,提出一种基于长短期记忆（LSTM）神经网络的S7协议模糊测试用例生成方法。将S7协议中的特征值字段分为可变字段和不可变字段,对可变字段进行模糊处理,对不可变字段做固定值操作,进而利用局部模糊实现对S7协议帧各字段的模糊分析,生成有效的测试用例。经过学习,模型可以提取到西门子S7协议的特征,自动产生满足协议结构的测试用例。实验对不同字段进行局部模糊,结果表明,该方法预测出的数据具备真实测试用例的特征,可生成大量对特征字段关联性较大的有效测试用例,提高代码覆盖率。     

南方电网云数据平台的架构设计方法

文章在综述中国南方电网目前信息化建设中存在的若干问题的基础上,重点分析了其数据平台在IAAS、PAAS及SAAS层面的需求,以及运行于该云数据平台之上的主要试点应用,提出了云数据平台IAAS、PAAS及SAAS层面的总体设计架构,并分析了其特点和优势。     

A hybrid approach for data clustering based on modified cohort intelligence and K-means

Clustering is an important and popular technique in data mining. It partitions a set of objects in such a manner that objects in the same clusters are more similar to each another than objects in the different cluster according to certain predefined criteria. K-means is simple yet an efficient method used in data clustering. However, K-means has a tendency to converge to local optima and depends on initial value of cluster centers. In the past, many heuristic algorithms have been introduced to overcome this local optima problem. Nevertheless, these algorithms too suffer several short-comings. In this paper, we present an efficient hybrid evolutionary data clustering algorithm referred to as K-MCI, whereby, we combine K-means with modified cohort intelligence. Our proposed algorithm is tested on several standard data sets from UCI Machine Learning Repository and its performance is compared with other well-known algorithms such as K-means, K-means++, cohort intelligence (CI), modified cohort intelligence (MCI), genetic algorithm (GA), simulated annealing (SA), tabu search (TS), ant colony optimization (ACO), honey bee mating optimization (HBMO) and particle swarm optimization (PSO). The simulation results are very promising in the terms of quality of solution and convergence speed of algorithm.