摘要
对大数据云存储下对海量数据传输的节能优化的,能够有效改善云存储过程高能耗现象。对海量数据传输的节能优化,首先需要对数据进行低负荷传输处理,对海量数据进行并行特征划分,完成海量数据传输的节能优化。传统方法将数据传输网域划分为多个虚拟环形结构,获取最小网络能量消耗量的价值成本,但忽略了对数据特征的并行预划分,导致数据传输过程能耗变化不大。提出面向大数据云存储的海量数据传输节能优化方法。依据合并小数据负荷编码块,分解海量数据负荷编码块,实现海量数据的特征并行划分,对低负荷传输中的指纹梯度进行优化处理,将此梯度引入到海量数据传输中,完成海量数据传输节能优化。实验结果表明,所提方法在数据传输请求总量不断增加的情况下,仍然具有低负荷性能。
The energy-saving optimization of massive data transmission in big data cloud storage can effectively improve the high energy consumption in the process of cloud storage. For energy saving optimization of massive data transmission, traditional methods divide data transmission domain into many virtual ring structures, but ignore the parallel pre-classification of data feature, resulting in little change of energy consumption in process of data transmission. This paper puts forward an energy-saving optimization method of massive data transmission for big data cloud storage. Combined with small data load code-block, we decompose massive data load code-block, and realize the parallel division of massive data feature. Moreover, we optimize the fingerprint gradient in low load transmission, and introduce this gradient into massive data transmission, thus complete the energy saving optimization of massive data transmission. Simulation results prove that the proposed method still has low load performance with the increase of total amount of data transmission requests.
作者
丁穗娟
DING Sui-juan(Guangxi Medical University, Guangxi Nanning 530021, Chin)
出处
《计算机仿真》
北大核心
2018年第5期160-163,共4页
Computer Simulation
基金
广西现代远程教育研究中心资助项目(2017GXOUWT12)
关键词
大数据
云存储
海量数据
节能优化
Big data
Cloud storage
Massive data
Energy saving optimization