- If Data Mapper needs to process large file (more than 50 MB), then always use splitting inside Data Mapper activity. You can enable splitting in Advanced Properties of Data Mapper activity and set Splitter/Merger x-path inside Data Mapper Applet. For more details on how to apply splitting in Data Mapper, please refer to Adeptia User manual.
- Always try to implement database caching inside Data Mapper. This can be achieved by making a DBQuery at Root node, storing the results in global variable and then for each record in input, using the global variable value to get the results from database. It would reduce the number of database interactions, probably one for whole Data Mapping and hence optimizing the performance of Data Mapper.
- There may be a need to increase the Operating System configuration for maximum allowed open file handles. Please check the help for your OS to see how to increase this parameter.
- If large number of concurrent process flows are executing in Adeptia Server, then make sure that number of concurrent connections to backend and log database are set to appropriate values. You can configure this in etc/server-configure.properties file.
...
abpm.jdo.maxActive (for backend database)
abpm.logdb.pool.maxActive (for log database)
You need to re-start Adeptia Services after changing these properties
- Running large number of concurrent process flows require more memory to be assigned to Adeptia Kernel. You need to assign sufficient memory to Adeptia Kernel so that it can execute the process flows without any hassle. To configure memory assigned to Adeptia Kernel, you need to change following in etc/launcher.properties file:
#Kernel JVM parameters
KN:JVM: -Xms512M –Xmx1024M -Xrs -XX:PermSize=64M -XX:MaxPermSize=128M
-Xms represents initial memory assigned to Adeptia Kernel at start up
-Xmx represents the maximum memory that can be assigned to Adeptia Kernel (if available)
You need to re-start Adeptia Services after making this change.
- To reduce logging on LIVE systems, all the process flows shall be deployed with "ERROR" logging level.
- The number of objects and processes within the backend database governs the performance of the user interface.
Therefore, when the backend database becomes large, you may need to increase the memory setting to this database:
- Administer > Setup > Application Settings > Update System Properties> Embedded Database Settings.
- Increase the abpm.embedded.indigo.memory value.
As well as the Adeptia WebRunner within etc/launcher.properties file:
#Webrunner JVM parameters
WB:JVM: -Xms128M -Xmx512M -Xrs -XX:PermSize=64M -
XX:MaxPermSize=128M
- To reduce the number of repositories inside Adeptia Server, all the process flows shall be deployed with Repository File Retention as "DELETE ON SUCCESS".
- Inside Adeptia Server, data and log clean up should be enabled and properly configured. The retain time (number of days for which data and logs would be retained) shall be properly configured according to the requirement. Please note that data clean up means cleaning of repository files only.
- For data clean up, retain time (in days) can be configured through Adeptia GUI at Administer > Configure > Application Settings > Update System Properties. Path to the property is: Maintenance/ Data Cleanup Properties. The name of the property is "abpm.appmanagement.retainTime" .
...
Links on this page direct you to the list of best practices that make your journey with Adeptia smooth and easy. These best practices not only guide you on how to use Adeptia for optimum output, but also help you fine-tune the performance of Adeptia in case of heavy data exchange.
Recommendations for naming an activity.
Configuration settings for higher performance of process flow.
Recommendations for naming Process Designer and design conventions.
Points to configure data mapping for higher performance.
Configuration settings for data cleanup.