Validation in OLTP Applications

In an online transaction processing (OLTP) application, all validation occurs on the server in batch mode.

In an online transaction processing (OLTP) application, clients must submit data in units of complete transactions to the server. If a client fails to return any data, or fails to complete processing of data, or returns partially nonvalid data, the server does not commit any of the data. Client and server processes run asynchronously in an OLTP application.

OLTP applications are also known as stateless applications. Web applications generally conform to OLTP architecture, because the HTTP protocol is stateless.

The key features of validation in an OLTP application are:

All validation must be conducted on the server, regardless of validation that may or may not be conducted by clients.
For example, you can implement sophisticated declarative and procedural validation in a web page using JavaScript, but the server page receiving such data has no way of knowing whether the client browser supports JavaScript, or if the client manually disabled scripts. Therefore, the server page must conduct validation checks itself.
Validation occurs in batch mode, with the server application processing many occurrences as a single transaction. No data can be stored until all occurrences validate.
It is important to ensure that the data supplied by the client is valid before it is submitted to the server. Therefore, declarative checks done by the client are still useful in an OLTP architecture. The server can also pass dynamic constraints to the client by parameters.
Certain functions and statements are not applicable to an OLTP architecture.
validate statements and validation functions that require database I/O are not suitable for use in a pure OLTP architecture.