To explain DAGs, vertices, and tasks, consider how Hive SQL queries are compiled and converted into a Tez execution graph also known as a DAG. A DAG is a collection of vertices where each vertex executes a fragment of the query or script. Directed connections between vertices determine the order in which they are executed. For example, the vertex to read a table must be run before a filter can be applied to the rows of that table.
As another example, consider when a vertex reads a user table. This table can be very large and distributed across multiple computers and multiple racks. Reading the table is achieved by running many tasks in parallel. The following figure shows the execution of a SQL query in Hive:
