SQL/MX 3.1 Installation and Management Guide (H06.23+, J06.12+)

Reorganizing SQL/MX Tables and Maintaining Data
HP NonStop SQL/MX Release 3.1 Installation and Management Guide663852-001
10-27
Recommended Practices for Improving import
Performance
If the table is not empty or has dependent indexes, import continues with the normal
load operation using TMF transactions.
For more information about
import command features and options, see the SQL/MX
Reference Manual.
Recommended Practices for Improving import Performance
This subsection describes recommended practices for improving the performance of
import operations on SQL/MX tables:
Running import on Empty Tables on page 10-27
Running import on Populated Tables on page 10-28
Managing Partitions to Improve import Performance on page 10-28
Managing Constraints to Improve import Performance on page 10-29
Managing Indexes to Improve import Performance on page 10-30
Managing Triggers to Improve import Performance on page 10-30
Managing Data Types to Improve import Performance on page 10-30
Running import on Empty Tables
You can import data into a table most quickly by using the fast loading technique,
which
import automatically uses when the table meets all these conditions and no
transaction size is specified on the
import command line:
It is empty.
It has no indexes.
It has no droppable primary key, unique key, or foreign key constraints.
It has no enabled triggers.
To further improve the performance of the fast loading technique,
import turns off the
audit attribute for the entire table at the start of the operation and turns it back on when
the operation ends.
If, while you are performing an
import operation with the fast loading technique,
another
import operation is attempted on the same table, the second operation fails
with a concurrent access error.
To improve the performance of an
import operation on an empty table:
Verify that all data being imported is contained in a single flat input file. OSS flat
files cannot exceed 2 GB. If the input file exceeds 2 GB, consider using
DataLoader/MX, which supports larger Guardian files. For information about using
DataLoader/MX with
import to load and maintain SQL/MP and SQL/MX
databases, see the
DataLoader/MX Reference Manual.
For an empty range-partitioned table: