Unlike single-byte character sets, Unicode can intermix different character sets (e.g. German with Arabic).

Unicode can be implemented by different character encodings, e.g. UTF-8, UTF-16 or UCS-2. Depending on the encoding, one character may take from 1 up to 4 bytes.

Example:

• Windows 1252: 1 byte per character

  

• UTF-8: 1 to 4 bytes per character

  

• UTF-16: 2 or 4 bytes per character

  

• UCS-2: 2 bytes per character

  

  UCS-2 encoding in this example is the same as UTF-16 but as it is limited to 2 bytes only, it can not display all characters of UTF-16. (UCS-2 is limited to the 'Basic multilingual plane' - the main Unicode code area.)

When Unicode is beneficial or necessary:

• Use of multiple languages from different character sets
• Use of Asian languages

The conversion of database columns to Unicode data types may cause:

• Database size increase by 1/3 to 1/2 of the original size.
• The DB engine and JDBC driver may have some overhead with Unicode data types (e.g. searching and sorting, transfer of data between the database and application)
• Problems with possible exceeding the limit on the total length of columns used in index (on MS SQL)