基于HANA数据模型的企业数据分析方法  

Analysis of enterprises data based on HANA data model

作　　者：李林 颜培俊 胡晓东 陈杰 袁一铭 LI Lin;YAN Pei-jun;HU Xiao-dong;CHEN Jie;YUAN Yi-ming(Jiangsu Electric Power Information Technology Co.,Ltd.,Nanjing 210024,China)

机构地区：[1]江苏电力信息技术有限公司,南京210024

出　　处：《信息技术》2022年第12期149-155,共7页Information Technology

摘　　要：针对海量企业数据应用困难的问题,基于采用下沉分析逻辑、并列执行等处理技术设计了新型的企业数据分析方法,该方法以HANA平台为基础,构建了混沌差分进化算法模型,该模型具有智能化、自动化对内存计算过程中的各种节点寻优的功能,克服了常规技术中存在的并行计算时,数据节点查找困难的问题。构建了新型的存储模块,该模块加入段式管理模块、页面管理模块和段式页面融合模块,能够实现不同方式的数据存储和数据存储方式转换。试验表明,该研究方法寻优能力较好,误差较低,数据存储量较大。Aiming at the problem of difficult application of massive enterprise data,a new type of enterprise data analysis method is designed based on processing technologies such as sinking analysis logic and parallel execution.This method is based on the HANA platform and constructs a chaotic differential evolution algorithm model,which is intelligent The function of optimizing various nodes in the process of memory computing can be optimized and automated,which overcomes the problem of difficulty in finding data nodes during parallel computing in conventional technologies.And build a new type of storage module,which adds a segment management module,a page management module,and a segment page fusion module,which can realize different ways of data storage and data storage mode conversion.Experiments show that this research method has good optimization ability,low error,and large data storage capacity.

关 键 词：企业数据 HANA 混沌差分进化算法模型 存储模块 融合模块 

分 类 号：TM721[电气工程—电力系统及自动化]