PDQ refers to the techniques that the database server can use to distribute the execution of a single query over several processors and coservers in Extended Parallel Server.
The database server can also use PDQ for queries that consume large quantities of non-CPU resources, in particular large quantities of memory and many disk scans.
A query that is processed with PDQ techniques is called a PDQ query. When the database server processes a PDQ query, it first divides the query into subplans. The database server then allocates the subplans to a number of threads that process the subplans in parallel. Because each subplan represents a smaller amount of processing time when compared to the original query, and because each subplan is processed simultaneously with all other subplans, the database server can drastically reduce the time that is required to process the query. Figure 74 illustrates this concept.