摘要
现有的数据存储方法在压缩数据时无法获取准确的密度峰值,导致数据负载均衡性较差,为提高数据分段均衡性,基于密度划分算法设计数字电网分布式数据存储方法。建立节点存储数据索引指针,通过节点递增顺序确定数据元区间;计算数据索引区域面积,计算面积与向量乘积;基于密度划分算法获取数据压缩密度峰值,以目标函数计算数据流目标长度;建立数字电网分布式数据存储模型,计算分布式数据库不同层级误差。在对比实验中,测试不同数据存储方法下分段数据负载均衡性。实验结果表明,将数据分为3-6段分别存储,只有密度划分算法在分段存储过程中,各数据节点数据量分割最均匀,数据负载均衡性最好。
The existing data storage methods can not obtain accurate density peak when compressing data,resulting in poor data load balance.In order to improve data segmentation balance,a distributed data storage method for digital power grid is designed based on density division algorithm.Establish a node storage data index pointer,and determine the data element interval through the node increasing order;Calculate the area of the data index area,and calculate the product of the area and the vector;The peak value of data compression density is obtained based on the density partition algorithm,and the target length of data flow is calculated by the objective function;The distributed data storage model of digital power grid is established to calculate the errors of different levels of distributed database.In the comparative experiment,the load balance of segmented data under different data storage methods is tested.The experimental results show that the data is divided into 3-6 segments and stored separately.Only the density partition algorithm has the most uniform data segmentation and the best data load balance in the segmented storage process.
作者
颜清
曹璐
李金讯
张蓓蕾
王鹏
YAN Qing;CAO Lu;LI Jinxun;ZHANG Beilei;WANG Peng(information and communication branch of Hainan Power Grid Co.,Ltd.,Haikou 570203,China;Hainan Power Grid Co.,Ltd.,Haikou 570203,China;China Southern Power Grid Hainan Digital Power Grid Research Institute Co.,Ltd.,Haikou 570203,China)
出处
《自动化与仪器仪表》
2022年第7期185-188,共4页
Automation & Instrumentation
基金
信通分公司2020年个性化专业级(域)运营管控应用(检修及通信集约化管理等)建设开发项目(729002020030102YX00058)。
关键词
密度划分
数字电网
分布式数据
数据存储
密度峰值
density division
digital power grid
distributed data
data storage
peak density
