Background thead to process data

Background processing of incoming data There is a function saveUserData in the web service, which receives and stores data from the user. Current version do just-in-time processing of incoming data. This is not good in terms of performance, under heavy load it is better to have a shorter transaction and response to the user as fast as possible. To get this, we will separate the process of storing and processing data. We will developed a single background process or thread that will handle the incoming data. It sould be priority handler with lower priority then base thread of web service. Better if it can be configured with system features or built in Java framework functions. If it cannot be done this way it should be provided with the calculation of time tuning of inactivity through the execution of standard features - a request to the database for the new batch of data. If it is growing - the handler works more rarely if falls – more often. Synchronization should be done using the database. That is, the main thread of web service writes the data to the database (this function is already exists), the second thread - data handler retrieves data from the database and process it. Data processing function already exists too. In the database table there is a field with a date-time of this data record was created. But it is better to use the ID, because the index by ID of anyone there, and an additional index will make insert data process slowly. In project used PostreSql, in which the ID later inserted data is greater then previously inserted. As a result, the process handler will be enough just to store ID of the last record processed. This ID is needed somewhere to store in a database table, what would stop at the service handler, this value is not lost. Several functions to process of incoming data should be done. Handlers should switch execution by “format” input value. Architecture of program code should be done such way, that each data format should be processed with separate class . Server infrastructure: Java, JBoss, PostgreSql, Hibenate