HP Neoview Release 2.4 Software - Manual (Page 89)

Glossary
cardinality Estimated number of rows that will be returned by the current operator, based on table statistics,
predicate values, and join criteria (if any) contained in the query. Represents the rows returned
from the FILE_SCAN operator. Accurate statistics are necessary for obtaining reasonable
cardinality estimates.
data source A logical name that defines the information required to access data. On the client side, it defines
the driver name, network address, and specific attributes, such as catalog and schema names.
disk process An instance of the Encapsulated SQL Access Manager (ESAM). A separate disk process manages
each volume.
Encapsulated
SQL Access
Manager (ESAM) The component of the Neoview platform that provides access to data. A separate disk process
manages each volume.
ESP parallelism Any parallel plan with at least one ESP plan fragment. ESP parallelism occurs when a plan
fragment executes within a special process called the executor server process (ESP).
execution plan See query plan.
executor server
process (ESP) A process that executes a portion of a parallel execution plan. ESPs can be used to perform sorts
or aggregation functions in parallel, or can be used to redistribute data to other ESP processes
in preparation for other parallel operations, such as joins.
hash join A join algorithm that creates a hash table for the inner table and joins the outer table by hashing
each other row and looking for matches in the hash table.
hybrid hash join A join method where the inner table is hashed on the join column into the memory of the process
performing the join. The outer table is read and the join column is hashed and matched against
the in-memory hash table. If the inner table is too large, it is subject to overflow to disk. The
original row order is not guaranteed unless ordered hash joins are used.
independent
parallelism An inherent feature of the Neoview database where two or more operators can execute
simultaneously and independently. Also called operator parallelism.
join A database operation that combines two or more tables into a single logical table so that the
data can be selected from all the tables at once.
join predicate A predicate that identifies and compares columns in a join operation.
master executor An process called the MXOSRVR. Controls the overall execution of the query. Also called the
root process or master process.
MDAM MultiDimensional Access Method. An optimized scan method that provides improved
performance for queries that contain missing predicates on key columns.
menu bar A list of commands at the top of the main window. The menu bar commands are Workspace,
Edit, Tools, Display, Windows, and Help.
merge join A join method where both tables are required to be sorted on the join column. A buffer of rows
is read from the inner and outer tables, and a row from the outer table is used to match inner
table rows in a match-merge pattern.
Neoview
platform A data warehousing platform that includes processors, storage, relational database software,
data services software, and professional services dedicated to supporting business intelligence
(BI) solutions.
nested join A join method that compares each row from the outer table with each row from the inner table
to find a match between the specified join columns from the two tables.
operator Represents an event in a query plan. Also called a node.
operator tree A structure that represents operators use in a query plan as nodes, with at most one parent
node for each node in the tree, and with only one root node.
optimizer A component of the SQL compiler that chooses query plans for retrieving data from a database.
parallelism Parallel execution in multiple disk process fragment instances, which is characterized by no-wait
communication (asynchronous access).
89