DataLoader/MX Reference Manual (G06.24+)

DataLoader/MX Examples
DataLoader/MX Reference Manual525872-002
6-10
Multiple Source Parallel Maintenance
These commands and their descriptions describe one way to accomplish the scenario
in Figure 6-4.
Command lines 1 through 2 execute a version of the DataLoader/MX process called
dbl0 (created using the build.sh script). In db10, the default user exit BUILDKEY is
replaced by one that, given an input record, returns its key. This strategy is necessary
because the DataLoader/MX process uses this key to determine which downstream
DataLoader/MX process receives the record:
1$ run dbl0 -E=err01 -I=infile1 "-O=kr(keyrange)" -S=100000 &
...
2$ run dbl0 -E=err0m -I=infilem "-O=kr(keyrange)" -S=100000 &
The preceding commands use these parameters:
-E specifies that execution time errors are recorded in the appropriate error listing
file.
-I directs the DataLoader/MX process to read the input from the appropriate input
file.
-O specifies that the output is written to the file kr (which is interpreted as a
keyrange file).
Figure 6-4. Multiple Source Parallel Maintenance
Note. gpri and cpu parameters are omitted from the example. The initial DataLoader/MX
processes should run at high priority, and the downstream DataLoader/MX processes should
be distributed across processors.
Data
Source
1
Data
Source
M
Part1
Part2
Part3
PartN
$dbl0 - initial
DataLoader/MX
1
$dblm - Initial
DataLoader/MX
M
DBL1
DataLoader/MX
N
DBL1
DataLoader/MX
3
DBL1
DataLoader/MX
2
DBL1
DataLoader/MX
1
VST013.vsd