A distributed file system and data processing method
A distributed file and data technology, applied in the field of distributed systems, can solve problems such as the extension of data consistency at startup
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0091] like figure 1 As shown, the distributed file system architecture information interaction diagram provided in the embodiment of the application introduces a distributed cluster composed of Zookeeper (consistency node), and the Zookeeper runs on the Namenode. In the embodiment of the application, the Namenode and the The ratio of Zookeeper is set to a one-to-one correspondence. Of course, it is not necessarily a one-to-one correspondence. The one-to-one correspondence is used here to better maximize the use of resources. This relationship is not fixed. The ratio can be set according to the actual situation. In theory, the larger the number of Zookeeper clusters, the better the performance, but this performance improvement is insignificant compared to the server hardware used more often, that is, the cost performance is actually very low. Setting the ratio of Namenode and Zookeeper to a one-to-one correspondence is a better cost-effective model.
[0092] The Master Namen...
Embodiment 2
[0112] like image 3 As shown, it is a structural diagram of a distributed file system provided by the embodiment of the present application. The system includes:
[0113] A Master Namenode (main control node) 201, for the node responsible for data writing
[0114] At least one Slave Namenode (from the control node) 202, for the node responsible for data reading;
[0115] An Observer (neutral control node) 203, a node for merging data mirroring and data logs;
[0116] The first detection and acquisition unit 204 is used to obtain the checkpoint corresponding to the time T from the neutral control node Observer when detecting that there is a newly added control node as the Slave Namenode in the distributed system at the time T. The data in it is obtained from the Master Namenode before the time T; the checkpoint is based on the regular merger of the data mirror FsImage and the data log Editlog; the time interval of the merger can be set according to the specific application, ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 