Open AccessResearch on Replica Placement Strategy for New Building Intelligent Platform
Author(s) -
Ying Yang,
Jianpeng Zhu,
R Zhang,
Xiaoliang Zhu,
Mei Wang,
C Wang
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1880/1/012037
Subject(s) - replica , backup , computer science , distributed computing , node (physics) , scheme (mathematics) , distributed data store , computer network , real time computing , operating system , engineering , art , mathematical analysis , mathematics , structural engineering , visual arts
The new building intelligent platform adopts single point storage in the data storage mode, which does not reflect the distributed characteristics of the platform. Once the node is damaged, the data will be lost. To solve this problem, this paper improved the data storage mode of the new building intelligent platform, combined Hadoop distributed file system (HDFS) with the platform, and realized the reliable backup of the platform data. For the deficiency of HDFS default replica placement strategy, the paper advanced a algorithm of replica placement for the new building intelligent platform based on PSO, and obtained the optimized replica placement strategy by considering the storage space residual rate of nodes, bandwidth residual rate and network distance between nodes. Simulation was on the CloudSim platform. The results of experiment show that the proposed scheme was superior to the default replica placement strategy in load balancing, which can reduce the load pressure of single node to a certain extent and improve the load balancing of the whole platform.