Fast predicate table scans using an architecture with a single instruction and multiple data

Provided is an approach in which a processor receives a search request for searching for data contained in a table data. The processor selects a column in the data table corresponding to the search request, and retrieves column data items from the selected column. In addition, the processor determines the width of the selected column and selects a search algorithm based on the column width determined from. Thereafter, the processor loads the column data entries in column data vectors and calculates search results from the column data vectors using the selected search algorithm.