【英文】Hadoop NameNode 合并规则

Posted on Edited on Views:

Preface

In Hadoop, when the NameNode receives requests for data operations, it records the requests in a log file and does not execute them immediately. Instead, when certain conditions are met, it sends this log file and the file system mapped in the disk to the SecondaryNameNode for merging operations.

Requirements to fulfill

  • Any of the following operations can trigger the merging operation:
  1. When the server restarts.
  2. When the log file size reaches or exceeds 64MB.
  3. When one hour has passed since the last merge.

Completion