An access path specifies the path chosen by a database management system to retrieve the requested tuples from a relation. An access path may be either (i) a sequential scan of the data file or (ii) an index scan with a matching selection condition when there are indexes that match the selection conditions in the query. In general, an index matches a selection condition, if the index can be used to retrieve all tuples that satisfy the condition.
Access paths are the alternative ways for retrieving specific tuples from a relation. Typically, there is more than one way to retrieve tuples because of the availability of indexes and the potential presence of conditions specified in the query for selecting the tuples. Typical access methods include sequential access of unordered data files (heaps) as well as various kinds of indexes. All commercial database systems implement heaps and B+ tree indexes. Most of them also support hash indexes for equality conditions.
To choose an access path, the optimizer first determines which matching access paths are available by examining the conditions specified by the query. Then, it estimates the selectivity of each access path using any available statistics for the index and data file. The selectivity of an access path is the number of pages (both index and data pages) accessed when the specific access path is used to retrieve the requested tuples. The access path having the smallest selectivity is called the most selective access path. Clearly, using the most selective access path minimizes the cost of data retrieval. Additional information can be found in .
- 1.Selinger PG, Astrahan MM, Chamberlin DD, Lorie RA, Price TG. Access path selection in a relational database management system. In: Proceeding of the ACM SIGMOD International Conference on Management of Data; 1979. p. 23–34.Google Scholar