system structure
In the previous article "Write Your Own Cloud Disk" - Establishing Requirements blog post, the functions to be implemented are proposed. In view of these functions, the following architecture of the system can be determined:
Among them, the HDFS distributed file system is used as the underlying file system, and the HBase distributed database is built on top of it. At the same time, HBase relies on Zookeeper for collaborative operations, thus building the Hadoop ecosystem. Hadoop has the characteristics of easy horizontal expansion, strong robustness and high reliability.
The Web system is not directly connected to HBase , but the method of operating HBase is registered in Dubbo to provide services. The Web system calls the services in Dubbo , which enables good service governance and decoupling.
Technical selection
Through the above architecture diagram, the technologies that need to be used can be listed:
(1) HDFS , as the underlying distributed file system
(2) HBase, as a distributed database, storing files in HBase
(3) Zookeeper as HBase collaboration Homework, at the same time as Dubbo's registry
(4) Dubbo, service governance and decoupling
(5) SpringMVC + Spring + Mybatis + Velocity , as a combination of Web system development
(6) MySQL, relational database
(7) Bootstrap , front-end frame
Pay attention to WeChat public account
Personal homepage original text: http://www.tongtongxue.com/archives/4556.html