0 votes
15 views
ago by (30.6k points)
We enabled Business Automation Insights on the Cloud Pak and see process dashboards. How does the data get there, and how do we get our business fields (order amount, region) into the dashboards?

1 Answer

0 votes
ago by (30.6k points)

Flow of the data:

  1. The workflow engine's event emitter sends events to Kafka topics whenever a process starts / ends, an activity changes state, a task is created / claimed / completed, and when a tracking point is hit (the business event emitter settings, in the CR on CP4BA and in 100Custom.xml on traditional BAW).
  2. BAI's Flink jobs read the topics, enrich and aggregate the events, and write them to OpenSearch / Elasticsearch indexes (one per event type: process summaries, activity summaries, human task summaries, tracking group events), optionally landing raw events in HDFS / object storage.
  3. Business Performance Center (and Kibana on older levels) visualise the indexes: OOB process and task dashboards, plus your own.

Your business fields travel through tracking groups, the same artifact that feeds the PDW: define a tracking group with fields (orderAmount, region), add tracking points in the BPD (or auto-tracking), map the variables; every tracked event carries the field values into BAI's tracking group index, where you build a dashboard (sum of orderAmount by region per week, time between two tracking points as a KPI).

# check the pipeline
oc get pods -n cp4ba | grep -E "bai|flink|kafka|opensearch"
# events arriving? (Kafka topic of the workflow emitter)
oc exec <kafka-pod> -- bin/kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic icp4ba-bai-ingress --max-messages 3
# index of the tracking group events
curl -sk -u elastic:$PW "https://<opensearch-route>/_cat/indices?v" | grep -i track

Tips: name tracking fields and tracking points for analysts (they become column names); keep personal data out of tracked fields; a tracking group belongs to a toolkit when several apps share the KPI; BAI keeps history independent of instance deletion (the PDW does not once instances are purged); the BAI event emitter can also send events to an external Kafka for your own data lake. On traditional BAW (WebSphere) BAI is installed separately and connected the same way (Kafka on the BAI side, the emitter on the BAW side).

References

Related questions

723 questions

807 answers

98 comments

4.8k users

Join BPM Community Discord Channel

Welcome to BPM Tips Q&A, Community wiki/forum where you can ask questions and receive answers from other IBM BPM experts and members of the community. Users with 2000 points will automatically be promoted to expert level.
Created by Dosvak LLC
Our Youtube Channel
...