Oracle recommends choosing Unicode for its universality and compatibility with contemporary and future technologies and language requirements. Autonomous Database on dedicated Exadata infrastructure currently uses AL32UTF8 as the database character set and AL16UTF16 as the national character set, as recommended for Oracle Multitenant.

However, for on-premises customers, with existing applications (and databases) using other character sets, migrating to Unicode character set can be a convoluted process with complex data analysis to avoid data truncation and corruption due to replacement characters. For enterprise customers using Peoplesoft, for instance, the pre-requisite of converting their data to AL32UTF8 character set as part of their migration to an Autonomous Database is quite complex. See Character Set Migration and Support Note 788156.1 for more details regarding the complexity of the exercise.

Autonomous Database lets you choose a character set of your choice from a list of supported character sets while provisioning an Autonomous Database. The list of supported character sets currently includes all database character sets supported on ASCII-based platforms. You can get the complete list of supported character sets using ListAutonomousDatabaseCharacterSets API. The supported national character sets supported are AL16UTF16 and UTF8.

Even though you are allowed to choose any combination of database and national character sets for an Autonomous Database, the Autonomous Container Database continues to use AL32UTF8 as the database character set and AL16UTF16 as the national character set. Oracle supports different database and national character sets inside an Autonomous Container Database for each database.

For step-by-step instructions on selecting a character set while creating an Autonomous Database, see step 12 in Create a Dedicated Autonomous Database.