| Title |
HDFS efficiency storage strategy for big data in smart city |
| ID_Doc |
36562 |
| Authors |
Xiang, M; Jiang, YZ; Xia, Z; Xu, LZ; Huang, CM |
| Title |
HDFS efficiency storage strategy for big data in smart city |
| Year |
2019 |
| Published |
|
| DOI |
10.1109/cac48633.2019.8996571 |
| Abstract |
With the rapid development of smart city and artificial intelligence technology, massive city-related information has been increased. The traditional storage strategy for big data can easily lead to hot nodes, which makes it difficult to meet requirement of big data storage efficiency. A Hadoop Distributed File System (HDFS) storage strategy for data based on the response time of data nodes (DNs) is proposed. Four parameters of memory utilization, network distance, bandwidth utilization and CPU utilization of DNs arc token as evaluation indicators of the strategy. Then the master node evaluates the response time of each data node (DN) based on BP neural network. Finally, the DNs with shorter response time would be chosen for data storage. The simulation results show that the proposed strategy can realize the distributed storage for big data and avoid the emergence of hot nodes. Furthermore, the strategy can effectively improve the response time of DNs and the load balancing of cluster. |
| Author Keywords |
big data; IMES; smart city; BP neural network; response time; load balancing |
| Index Keywords |
Index Keywords |
| Document Type |
Other |
| Open Access |
Open Access |
| Source |
Conference Proceedings Citation Index - Science (CPCI-S) |
| EID |
WOS:000679040102079 |
| WoS Category |
Automation & Control Systems |
| Research Area |
Automation & Control Systems |
| PDF |
|