Dataloader/MP Reference Manual

Table Of Contents
Running DataLoader/MP
DataLoader/MP Reference Manual424148-003
3-10
Considerations—DataLoader/MP Process
filename [(feature [,feature]...])]
specifies the Guardian file name including node name, volume, and subvolume or
DEFINE name if necessary, or an OSS file name, followed by an optional list of
features enclosed in parentheses and separated by commas. If a file name
contains any “/” characters, NM DataLoader/MP will assume it is an OSS file name.
The order of the features is not important. There must not be any embedded
spaces in the entire construct. For example:
$data1.master.file1(indirect)
references the file $data1.master.file1 and requests the INDIRECT interpretation
(described on page 4-7).
$data1.master.file1(max=1000,8.63%)
references the file $data1.master.file1 and requests both the MAX and % modifiers
(described on pages 4-9 and 4-10). No spaces appear anywhere.
You can reference a physical file name by using a DEFINE. Normally, you give the
DEFINE name with an equal sign (=) preceding it. However, DataLoader/MP’s
arguments include an equal sign (=), and to TACL a statement such as
-I==definename
is interpreted as “-I” followed by a comment that includes the rest of the line. Use a
tilde (~) to let TACL know that the double equals sign (==) is not a metacharacter:
-I~==definename
This will tell TACL not to treat the double equals sign (==) as a comment. TACL will
then drop the tilde and pass all the other characters to DataLoader/MP.
This should be done only on the command line. When you use a DEFINE name in an
INDIRECT file, a KEYRANGE file, or other file, do not use a tilde.
Using the -P Parameter
The -P parameter is rarely used in a production data load but is very useful when you
initially set up a database. Suppose you have a large amount of data, perhaps
containing hundreds of tapes. You know it contains 500,000,000 records. You want to
partition this data over ten disks but have only a rough idea of the key values. To
partition the data, you must know the exact key values to divide this data into even
50,000,000 record partitions.
-P can provide these values. First run DataLoader/MP to get the boundaries, then set
up your database with the partition boundaries you have determined.
You could create a version of DataLoader/MP, db10, that replaces the default
BUILDKEY with one that looks at each record and determines its key, then run it with
this command:
1>db10 -I=infile -P=10
DataLoader/MP will display the partition boundaries in a report when it completes.