Strategy for handling oracle tables with number PK values > 2.2 billion

Did you try to map the data types in the XSD file? We work with number(38,0) in Oracle and map it to string in Appian.

Can you give an example of how you did the xsd mapping for a full CDT? When I create a CDT with a single field, the PK of the view which is NUMBER data type mapped to Appian Text type, the Data Store successfully verifies, and query works as you have stated. But as soon as I add another field to the CDT, I get the error for the previously working element: The data source schema does not match the type mappings: Wrong column type in [] for column []. Found: number, expected: varchar2(255 char). Thanks

Sure. You download the XSD, modify it and upload it again. Check docs.appian.com/.../Mapping_CDTs_to_Pre-defined_Database_Tables.html

<xsd:element name="billionId" nillable="true" type="xsd:string">
<xsd:annotation>
<xsd:appinfo source="appian.jpa">@Id @Column(name="BILLIONID", columnDefinition="NUMBER(38)")</xsd:appinfo>
</xsd:annotation>
</xsd:element>

This will not have any impact on other fields or cause issues when you add any new fields.

marky i agree with Stefan Helzle , you can try creating the Primary Key as String in DB Table and map the same using the CDT as Primary Key and uncheck AutoGenerate checkbox in CDT, now while inserting record into DB, get the totalCount of that table and use tostring(totalCount+1) as the Value for Primary Key, this can do your Job.

Now to get the totalCount, i would suggest you better go for the Smart Service instead of using QueryRule or QueryEntity, and you can execute the following Query: select count(*) from TableName; this will give you the totalCount for the mention table.

But i would like to know one thing, will you really be inserting this much huge data into a DB table because for processing this much of huge Data, normally we need to for BigData, Hadoop.

However this should work for you requirement