Tables can be created based on other tables. This implies that users have the freedom to use any given table or to combine any number of existing tables in order to create a new one.

This fact is very important in the context of a design process because the main purpose of it is to add value to the raw data extracted from the different data sources; and the only way to achieve such objective is by reformatting, cleaning and deriving new data from existing one.

This implies the creation of new tables by means of Direct Mapping, Joining or Merging existing tables; and the creation of new fields through mathematical, string and logical calculations.

A calculated field is normally the result of a combination of several fields through a given formula and hence it can be completely user-defined, having properties that are not necessary similar to those of any of its contituting fields.

Furthermore, a calculated field may also be defined only as a constant or variable expression without any constituting field at its basis.

For instance, using calculated fields it would be possible to determine whether a bill has been paid or not based on a payment date. If a payment date is not registered, the bill has not been paid or vice-versa. Other common calculations are the concatenation of "Name" and "Last Name", formatting of dates, calculation of YTD, etc.

For these purposes, Quella Creator offers a good selection of Mathematical, Logical, String and Date Functions.

A relation is a connection established between fields or columns belonging to two or more tables so that the latter can be combined. A relation works by comparing the content of the connected fields record per record and producing a resulting record only and only if the compared values in the given record are identical.

Relations between tables are defined by dragging a field from one table to the other. In this case, the relationship window shows an existing relation between "Purchase Head" and "Purchase Line" via "PO" (Purchase Order Number). It also shows that the defined relation is of type INNER JOIN. In general, relations can be defined to be of type INNER JOIN, LEFT OUTER JOIN or RIGHT OUTER JOIN.

At the bottom of the window, all the relations that are defined for a given table are shown. In the same window, it is also possible to edit and delete existing relations.

Table filters are a very efficient way to omit a particular record or group of records from a given table. With the help of filters, it is possible to easily define the content that can be allowed in a resulting table and in this way to define which records are to be extracted or transformed.

Filters can combine several filtering criteria based on the same or different fields. The number of filtering criteria that can be included in a table is in theory unlimited.

The main purpose of aggregations is to generate summary values for the tables where these aggregations are applied. When a table is aggregated all the values contained by its constituting fields are processed so that to produce a single resulting value for each field.

These resulting values may also be broken up into individual groups according to given fields if so defined in the aggregation. When an aggregation is applied to a table, all its fields must be part of it either by summarising or grouping their values.

An index is a structure that orders the values of a field in a table so that the ordered values can rapidly be found when required. To do so, the index provides pointers to the data stored in the indexed field and orders them according to a specified sort order. From a functional point of view, indexes in tables are similar to those in books. They allow the system to quickly find information in a table without the need to scan the entire table.

In terms of which fields to index, the best candidates are those directly involved in search operations. Such operations are Table Relationships, Filters and Aggregations through the Aggregation Function Group By.