Difference Between datastore , datatype and recordtype

A datatype defines a data structure which can be used in a datastore to persist data to a SQL database. Other use cases are data exchange with external systems or internal helper structures.

A datastore represents a set of datatypes and a connection to a SQL database. Used to manage the table schema.

A recordtype is used to represent business data to the user and allow interaction using processes as related actions.

Ok (deep breath)...here goes:

• a datatype simply conveys a "type" to the data. There are primitive data types (e.g. Text, Integer, Boolean). And there are complex types (e.g. Address) which can be assembled from a combination of primitive and complex types. Note that a complex datatype does not have to be used to map to a database table or view (i.e. you can use complex datatypes in you application that do not map to a database or view) but that is the most common use case
• a datastore is a "connection" to a database. You use it to host the complex datatypes that you want to map to your database entities (tables, views). A datastore is usually associated with a specific Appian application, and you can have different datastores connecting/mapping to the same database
• a Recordtype is an Appian object type used to surface special instances of data that you want to conduct processing against. a Recordtype can be mapped to a database entity (so called "entity-backed") or to a rule that returns some data (so called "service-backed" as is often used to connect to a web-service) or to some in-flight instances of processes in Appian (so called "process-backed"). The key thing about Records are that they represent key data entities that you want to expose to end-users and which can be searched for, opened up to view, and to run processing against.