Hierarchy
EA-RETAIL (Software Component) SAP Enterprise Extension Retail⤷
LO-MD-QO (Application Component) Quantity optimizing⤷
WLC_PARALLEL_CUST (Package) Retail: Load Control and External Parallel Proc./Customizing
IMG Activity
| ID | W_GE_0500 | Settings for External Parallel Processing |
| Transaction Code | S_ALN_01000254 | (empty) |
| Created on | 20011127 | |
| Customizing Attributes | W_GE_0500 | Settings for External Parallel Processing |
| Customizing Activity | W_GE_0500 | Settings for External Parallel Processing |
Document
| Document Class | SIMG | Hypertext: Object Class - Class to which a document belongs. |
| Document Name | W_GE_0500 |
Use
In this IMG activity you make settings to control external parallel processing. External parallel processing allows you to convert existing reports, which will not allow you to carry out any internal parallel processing in your worklist, into reports that automatically plan several online processes in parallel on the servers available in a server group.
For more information, see the report documentation:
Activities
In the development system, generate a separate report Rp (transaction WLCPAR) with the report RWLC_PROCESS_IN_PARALLEL for every report R that is to be parallel processed. Rp copies the preselection of the worklist, the distribution of the parallel jobs, the creation of selection variants Vrp (that are derived from the variant Vr) for the parallel jobs, and the start of the parallel jobs for report R.
If you want to generate a report like this, enter the following information:
For every report R that is to be parallel processed, you must enter in Customizing (transaction WLCPARC) in the development system how the worklist is to be determined. It is absolutely imperative that you have good knowledge of how the report R that is to be parallel processed actually works.
NOTE: Every time a change is made in the views Mandatory Options and Additional Selection Options, you have to start transaction WLCPAR for the reports concerned again so that the changes are included in the generated reports.
1. Which tables do you have to choose (preselected worklist)?
- Worklist determination from just one table
- In the view Mandatory Options in the Preselection field choose the value Generic single-level. In the Primary Table field, enter the table you want to select.
- Worklist determination from two tables
- In the view Mandatory Options in the Preselection field choose the value Generic two-level. In the Primary Table field, enter the first table you want to select. In the Secondary Table field, enter the second table you want to select. Generally, this involves header and entry tables. The entry table repeats the primary key of the header table.
- Worklist determination from more than two tables or tables without links
- In the view Mandatory Options in the Preselection field choose the value Using a Function Module. In the Function Module field, enter the name of the function module that selects the worklist.
- The function module WLC_PRESELECT_EXAMPLE provides you with an example in the system. In the export table ET_WORKLOAD, specify the values for the selection options that are used for distributing the workload among different jobs (see question 4). The same values or, indeed, different ones can be used for the distribution (see question 3). You can also specify the factor with which an entry workload is valuated.
Rp determines the selection options and parameters that relate to one of the two tables entered from the selection screen of report R. All the values are taken from the variant Vr and the worklist selected from the tables is therefore restricted.
Unfortunately the selection options and parameters are not always defined in such a way that the reference to the tables that are to be selected is produced immediately. However, it is important to be aware of these limitations so that Rp selects the same worklist as R. You can expand the display of these restrictions in Customizing in the view Additional Selection Options. The entries made there are converted into selection options and parameters on the selection screen of R. The values can be defined using the variant Vrp.
You can make the following entries in this Customizing view:
SelOpt/Param Technical name of the selection option or parameter
Description In the language requested
Sort Sequence Sequence (1-999) in which the selection options are displayed on the selection screen
Selection Type S = selection option, P = parameter
Table name Table whose selection is to be restricted
Field name The selection restriction is based on this field
The following optional values are defined with the variant Vrp. You can also define multiple conditions here. In Customizing you can only define single conditions.
Plus/minus sign Include or exclude values
Option Comparison operation
From Lower limit
To Upper limit
Example (1):
You want to be able to parallel process the posting of the documents. The documents are stored in the database in a header table TH and an entry table TI. You can only post the documents once they have a particular status that is contained in the status table TS.
TH has the primary key TH-K. TI has the primary keys TI-K and TI-NR. So the key of TH is copied and enhanced to include NR. Generally the same name and data type is used for K so that both tables can easily be linked. The key of TS could have another name: TS-L. For Rp it is sufficient if only the data types match to link the two tables.
Preselection should always be set up generically as a two-level selection. TS is selected first with the option of restricting the selection with the relevant status values and/or data area which can be stored in the report variant or as additional selection options for Rp. Finally TI is selected. This is possible because the data types in the keys of TH and TS at least match up and make up the initial part of the TI key.
2. How is the range of the worklist evaluated (distribution)?
The workload should be evenly distributed among the parallel jobs. To do this, the data collected in the first step is examined and aggregated if necessary.
Often the number of rows for each header table entry is a measure of the workload in two-level selections (for example, order items) or the frequency of a particular value (number of different materials). For this reason, the workload must be calculated cumulatively which means the number of rows is counted:
In the view Mandatory Options in the Workload Calculation (WorkloadCalc.) field, select the value Cumulative (Single Values are Added).
In single-level and in some two-level selections it is only the occurence of a particular value - regardless of its frequency - that affects the workload:
In the view Mandatory Options in the Workload Calculation (WorkloadCalc.) field, select the value Single Value (Only its Occurrence Matters).
In cases where the document workload depends on a data field, you can enter this data field in the view Mandatory Options in the Field for Weighting field. Depending on the type of preselection, you maintain the field behind the primary or secondary table. If the field type is not numerical, the content is ignored in later processing and the document is weighted with a 1.
Example (2):
Document posting takes longer the more items a document has. For this reason, the number of items for each header document is cumulated.
3. Which field should be used for distribution among parallel jobs?
The parallel jobs should contain a worklist that does not overlap. A value is taken from the worklist for this to group the jobs. Select the value so that there are no lock conflicts produced by the parallel processing.
In some reports this value is contained in the selection screen of R so you only need to choose the radio button in the selection screen of Rp ("Selection of values for grouping worklists"). The values are taken from the table field with which the selection option is defined.
You can make settings for this in Customizing (view Default Values for Selection Screen) in the field Select.Options for Job Creation. The value that is to be entered can be determined using the F1 help for the relevant radio button.
In other cases the value for grouping parallel jobs is not contained on the selection scre
Business Attributes
| ASAP Roadmap ID | 306 | Establish System Administration |
| Mandatory / Optional | 2 | Optional activity |
| Critical / Non-Critical | 1 | Critical |
| Country-Dependency | A | Valid for all countries |
Assigned Application Components
| Documentation Object Class | Documentation Object Name | Current line number | Application Component | Application Component Name |
|---|---|---|---|---|
| SIMG | W_GE_0500 | 0 | HLA0006433 | Logistics - General |
Maintenance Objects
| Maintenance object type | C | Customizing Object |
| Assigned objects | ||||||
|---|---|---|---|---|---|---|
| Customizing Object | Object Type | Transaction Code | Sub-object | Do not Summarize | Skip Subset Dialog Box | Description for multiple selections |
| VC_WLC_OPTIONS | C - View cluster | SM34 |
History
| Last changed by/on | SAP | 20011127 |
| SAP Release Created in | 110 |
