The Hash Table Scan icon indicates that the entire temporary hash table will be scanned and all of the entries contained with the hash table will be processed. All of the entries contained within the hash table will be processed, unless the icon label is Hash Table Scan Distinct or Hash Table Scan Distinct, Preload, in which case only one entry from each set of duplicate entries that would be selected within the hash table will be processed. A hash table scan can be chosen when the optimizer is considering a plan that requires the data values to be collated together but the sequence of the data is not required. The use of a hash table scan will allow the optimizer to generate a plan that can take advantage of any non-join selection while creating the temporary hash table. An additional benefit of using a hash table scan is that the data structure of the temporary hash table will usually cause the table data within the hash table to remain resident within main memory after creation thus reducing paging on the subsequent scan operations.
Also referred to as: Hash Table Scan, Preload; Hash Table Scan Distinct; Hash Table Scan Distinct, Preload