Based on Oracle, processing an English character takes 1 byte, but 3 bytes for a Chinese character.
For example:
select ename, length(ename), lengthb(ename), empno, length(empno),lengthb(empno)
from emp001;
ENAME LENGTH(ENAME) LENGTHB(ENAME) EMPNO LENGTH(EMPNO) LENGTHB(EMPNO)
-------------------- ------------- -------------- ---------- ------------- --------------
郭靖 2 6 100 3 3
黄蓉 2 6 200 3 3
裘千尺 3 9 300 3 3
洪七公 3 9 400 3 3
杨过 2 6 600 3 3
梅超风 3 9 500 3 3
From storage point view, deciding to use Chinese characters in the table values would cost 3 times more space
usage than its English counterpart. How much impact would have on the database performance
when the data activity on a table, such as insert and update data, is increased?
Can we compare apple and orange in this case?
It seems that to process Asia language, like Chinese, Japanese, and Korean, the character set in database system
will cost more than in English. And perhaps we can‘t comapre apple and orange in this case after all.
原文：https://www.cnblogs.com/zwang562020/p/13199262.html