Use the DB Alias tab to define the parameters needed for theTeradata
loader. The Optim™ solution
supports the Teradata loader for both FastLoad and MultiLoad. The
availability of certain options and fields might be dependent upon
your entries for related options and fields on the tab.
Load with Teradata Named Pipe
Load performance
might be improved by using the Teradata Named Pipe Access Module.
When this option is used, the load process runs in a single phase
rather than a two-stage process, resulting in reduced elapsed time
for load processing. Because the row data is loaded immediately after
conversion, space requirements are also reduced.
For Teradata, the Optim solution writes the data to the named
pipe and the Teradata access module reads and copies the data, which
is loaded. Optionally, you can write the data to be loaded to a fallback
file, so that you can restart the load with Teradata utilities. The
solution writes data to the fallback file, based on the save interval
you set. After the load process completes, the fallback file is automatically
deleted. The fallback file extension is .fbf. The file is in the %TEMP%
or %WINDIR%\temp directory.
Mode
- Insert
- Insert rows from the source file into empty destination
tables. If destination tables contain data, the loader returns an
error.
- Replace
- Clear all data in the destination tables and replace
with rows from the source file.
- Delete
- For MultiLoad only. Delete rows from the destination table. The
source and destination tables must have the same name, same creator
ID, and same DB alias in the table map.
Utility
- MultiLoad
- Generate a file to use with MultiLoad. If a table contains LOB
columns, Basic Teradata Query (BTEQ) is used instead of FastLoad or
MultiLoad, which do not support LOBs.
- FastLoad
- Generate a file to use with FastLoad. If a table contains LOB
columns, Basic Teradata Query (BTEQ) is used instead of FastLoad or
MultiLoad, which do not support LOBs.
File Type
- ASCII
- Generate an ASCII file of fixed-length rows.
If the data for a column is shorter than the maximum width of that
column, the data is padded with trailing blanks. This means more bytes
are sent over the network to the Teradata server, but a delimiter
is not required.
- ASCII Delimited
- Generate an ASCII delimited file type, which
is a variable-length format with a delimiter character separating
each column value. If you select this file type, select a delimiter
that does not exist in the column data. This file type can be loaded
to a database that is enabled with DPF (Data Partitioning Feature).
When character columns are processed, the solution pre-scans the data
and issues an error message if a delimiter is present in the text.
If the data includes free-form text, such as a description field,
it may be difficult to choose a delimiter that does not appear in
the data. You can select a control character, that is less likely
to be present in the data.
- Delimiter
- Eliminating a scan for delimiters in character
data columns can improve load performance. To eliminate a scan, select
one of the following. The presence of the selected delimiter in the
source data causes a conversion error during the load process.
X'01' (No Pre-scan)
X'02' (No Pre-scan)
X'03' (No Pre-scan)
X'04' (No Pre-scan)
X'05' (No Pre-scan)
Scan while generating the loader
file. Select any of the following values from the Delimiter list:( ) * / ; ? | < > =
X'FA'
X'FB'
X'FC'
X'FD'
X'FE'
X'FF'
- Binary
- Generate a file with values in binary representation. This format
uses FastLoad and MultiLoad capabilities to execute the load faster
and more efficiently than ASCII or ASCII Delimited.
Options
- Perform Load
- Run the loader automatically after file conversion
processing is complete. Clear this check box to prepare the data in
the appropriate format and create the BAT file to run the loader.
Before running the loader, you must edit the BAT file to include the
password information.
- Use Named Pipe
In a Windows or UNIX environment, load performance
might be improved by using a named pipe. When this option is used,
the load process runs in a single phase rather than a two-stage process,
resulting in reduced elapsed time for load processing. As the row
data is loaded immediately after conversion to a loader format, space
requirements are also reduced significantly, as compared to the two-stage
load process. The solution creates the named pipe and writes the
file to it. The load utility reads the data from the named pipe and
submits the data to the loader.
- Delete files if Successful
- Fast load confirmation is
not available. Delete the data files after the loader completes successfully.
This option is unavailable if you selected Use Named Pipe.
- Delete files if Failure
- Fast load confirmation is
not available. Delete the data files if the loader does not complete
successfully. This option is unavailable if you selected Use
Named Pipe.
- Load When Source is Empty
- Complete the load if a table to be loaded contains no rows, deleting
any rows from the target. If not selected, empty source tables are
skipped.
- Use In-line LOBs
- Include LOBs in a data file (inline with the table data).
- Enable delimiter pre-scan
- Scan for delimiter characters while building the loader file.
- Create Exception Tables
- Create exception tables for specified tables.
- Ignore Missing Rows
- For MultiLoad Delete only. If rows cannot be deleted because
they do not have unique keys, selecting this checkbox prevents the
rows from being written to an error table.
- Ignore Duplicate Rows
- For MultiLoad Insert and Replace only. If rows cannot be inserted
because they have duplicates, selecting this checkbox prevents the
rows from being written to an error table.
Checkpoints
Both MultiLoad and FastLoad
support the use of checkpoints. Checkpoints are entries posted to
a restart log table at regular intervals during the data transfer
operation. If processing stops while a job is running, you can restart
at the most recent checkpoint. For example, if FastLoad loads 1,000,000
records to a table and checkpoints occur every 50,000 records, an
entry to the restart log table occurs when 50,000 records are loaded.
If the job stops after loading record 60,000, you can restart the
job beginning with record 50,001.
- 0
- Default. Bypass the checkpoint function. Nullifies the MultiLoad
restart and reload capability but lets you use an INMOD routine that
does not support the checkpoint function
- 1 through 999999999
- For MultiLoad, the number of minutes between checkpoints. For
FastLoad, the number of records to process between checkpoints.
Error Limit
- 0
- Default. Stop processing when the first record is rejected.
- 1 through 999999999
- The error threshold for the number of rejected records that can
be written to the error tables during the acquisition phase of a load.
Max Sessions
The maximum number of load
sessions that can be logged on. Enter a number from 1 (default) to
200.
Work path for data files
The path for storing
temporary data files.
Additional Loader Parameters
You can augment the loader arguments
created automatically by the Optim solution
with additional loader parameters. The parameters you create are appended
to the list created automatically, but are not validated before starting
the loader. For valid operands, see the database loader documentation.
If additional loader parameters are forced from within product options,
you cannot modify them.