The runppmcompress argument

-ps <file name>

This argument is used to specify the XML file containing the paramset.

-mode <test|execute>

test: This argument enables you to perform a test run using the output statistics to identify which process instances and how many would be aggregated. This does not make any changes to the database.

execute: The process instances are aggregated. This operation cannot be undone. A statistics is output.

Default value: test

-excluding <n>[y|q|m|w|d|h]

The process instances lying in the specified time period are not aggregated, even if they are included in the filter. The period is specified as a number of particular sections of time. The start time of the program is used as the reference time.

Abbreviation

Time section

Example

y

Year

1y: Process instances from January 1st of the current year to the start time of the command line program are not deleted.

q

Quarter

2q: Process instances for the current last completed quarter are not deleted.

m

Month

 

w

Week

4w: Process instances for the current week and the last three weeks are not deleted.

d

Day

 

h

Hour

24h: Process instances for the current hour and the last 23 hours are not deleted.

Example

In the analysis component, the Number of processes measure has been displayed over time with the filter 1.1.2003 to 15.2.2003 with a weekly scale for the Other orders process type. The corresponding paramset was saved in the file paramset.xml:

<?xml version="1.0" encoding="ISO-8859-1"?>

<!DOCTYPE paramset SYSTEM "paramset.dtd">

<paramset>

<analysewizparam selectedrow="0"/>

<chartparam>

<chart-legend-visible> false</chart-legend-visible>

<chart-scale>1.0,10.00.10.0</chart-scale>

<chart-zoom>75</chart-zoom>

<chart-chartposition> 0.175,0.1</chart-chartposition>

<chart-colinfo/>

</chartparam>

<cpiparam/>

<epkparam/>

<favoritparam />

<kiquery>

<keyindicator>

<criterion name="PNUM">Number of processes</criterion>

</keyindicator>

<iteration>

<criterion name="TIME">Time</criterion>

<refinement name="BY_WEEK">By week</refinement>

</iteration>

<filter>

<criterion name="TIME">Time</criterion>

<filteritem operator="or">

<datarange operator="range">

1. Jan 03 - 15 Feb 03

<dataitem value="1.1.2003 0:00">

01.01.03 0:00

<datatype name="TIME">Time</datatype>

<scale name="MINUTESCALE" factor="60.0">

Minute(s)

</scale>

</dataitem>

<dataitem value="15.2.2003 23:59">

15.02.03 23:59

<datatype name="TIME">Time</datatype>

<scale name="MINUTESCALE" factor="60.0">

Minute(s)

</scale>

</dataitem>

</datarange>

</filteritem>

</filter>

<filter>

<criterion name="PROCESSTYPE">Process type</criterion>

<filteritem operator="or">

<dataitem>

Order processing\Other orders

<datatype name="TEXTPAIR">Text pair</datatype>

<scale name="LEVEL2SCALE" factor="2.0">

Detailed

</scale>

</dataitem>

</filteritem>

</filter>

</kiquery>

<navparam>

<nav-view>chart</nav-view>

</navparam>

<tableparam view="listtable">

<tbl-cross totalresult="yes"/>

<tbl-instance split="horizontal"/>

</tableparam>

</paramset>

Executing the command line

runppmcompress -user system -password manager -ps paramset.xml -mode execute

will aggregate the process instances in the period from 1.1.2003 to 15.2.2003 by week.

-recoveryfile {yes|no}

With -recoveryfile yes (default setting) you specify that the analysis server recovery files relevant for process analysis are to be created after successful aggregation under <installation directory>\ppmmashzone\server\bin\work\data_ppm\recovery\<client>.