A bucket is a precalculated value set that is stored either in a database table or in an index in SAP NetWeaver BW Accelerator and can be used as a filter in a BEx query.
The filter is defined in a BEx query as a variable of type Precalculated Value Set and executed at runtime. It is important here that the filter is executed on the database or in SAP NetWeaver BW Accelerator before all the data from the bucket has to be loaded to OLAP processing. This makes for significant runtime improvements if large amounts of data are processed in the bucket.
A variable that refers to buckets can only be used as a filter in characteristic restrictions. It cannot be used in default values, structure elements and the like.
To fill a bucket with data, you can either use a reference table or add the data directly.
You are in the Bucket Definition screen (transaction RSBUCKET).
To create a bucket, proceed as follows:
Input help is provided for existing buckets.
Note that displaying a large table can be time-consuming if a large number of rows is set as standard for it in the Data Browser (transaction SE16). You can also expect long runtimes when displaying data from a large index in SAP NetWeaver BW Accelerator. You do have the option however of refining the display and thus minimizing the number of values that are actually displayed.
This function is particularly useful when using InfoCubes in SAP NetWeaver BW Accelerator. When doing this, all of the data in the table has to be loaded to the working memory, meaning that the size is restricted here too. The name of the index that is created contains a prefix. This prefix is made up of the system name and a special namespace 0BW:RST:<SID>:<Name>, where <SID> is the system name and <Name> the name of the table.
The system has precalculated the required value set, which is now available in the form of a bucket. You can use this bucket in a BEx query to define a variable of type Precalculated Value Set (seeDetails).