Database Reference
In-Depth Information
The EXPLAIN function
The EXPLAIN function displays the query execution plan for the query. Every node
represents a single operation and it reads plans from bottom-up as each node feeds
rows into the nodes directly above it. The bottom nodes of a plan are usually se-
(redistribute, explicit redistribute, broadcast, or gather motions). These are the op-
erations responsible for moving rows between the segment instances during query
The EXPLAIN ANALYZE function is run on queries to identify any areas where the
query performance can be improved. This function always executes the query.
The following syntax should be used to run EXPLAIN ANALYZE ; we would need to
use the EXPLAIN ANALYZE explicitly in a transaction:
Dynamic Pipelining in Greenplum
Dynamic Pipelining framework enables parallel data flow both during distribution and
querying. Following are the key aspects of this feature:
• Combines high-speed UDP interconnect and a runtime execution environ-
ment for big data workloads
• Data from upstream components in the dynamic pipeline are transmitted to
downstream components through UDP interconnect
• Enables queries to run without materializing intermediate contents to disk
The following figure explains the function of this feature:
Search WWH ::

Custom Search