2022学习总结专用(***)

1.14

1.hadoop序列化总结:

  • 序列化是把内存中对象变成二进制序列的过程,目的是把易失的对象转化成永久保存的磁盘文件、或者进行网络传输。反序列化是从二进制序列读取对象的过程。
  • 对于复杂的组合数据,可以自定义bean实现其序列化,需要实现Writable接口的两个方法write和readfields。wirte就是将对象往麻袋里写,readfields就是从麻袋里往外掏数据,赋给空对象。

2.

————————

one point one four

1. Hadoop serialization summary:

  • Serialization is the process of turning objects in memory into binary sequences. The purpose is to convert volatile objects into permanently saved disk files or network transmission. Deserialization is the process of reading objects from binary sequences.
  • For complex composite data, you can customize the bean to realize its serialization. You need to implement the two methods write and readfields of the writable interface. Wirte is to write an object into a sack, and readfields is to take out data from the sack and assign it to an empty object.

two