Broadcom
ORSYP Dollar Universe → Apache Airflow
Migrate Dollar Universe (Univiewer) uprocs, sessions, and rules to Apache Airflow DAGs.
Dollar Universe (now part of Broadcom after the ORSYP acquisition) uses a unique model of uprocs (unit processes), sessions, and rules to define workloads. Its multi-node architecture and uproc chaining require careful decomposition before Airflow migration. BatchFoundry maps Dollar Universe constructs to Airflow equivalents, including session-level parallelism and uproc parameter passing.
Concept Mapping
Uproc (unit process)Task / Operator
SessionDAG
Rule (scheduling rule)schedule_interval / timetable
Uproc chainTask dependency chain (>>)
Node (execution host)Airflow worker queue
Variable (UNV_VAR)Airflow Variable / Param
Mutual exclusion (mutex)Airflow Pool (slots=1)
Promotion / areaAirflow environment / deployment
Alerton_failure_callback
Log (UNIVIEWER)Airflow task log
The Hard Parts We Handle
- Uproc parameter and variable resolution
- Session-level parallelism → Airflow concurrency settings
- Multi-area promotion logic
- Dollar Universe mutex → Airflow Pool with 1 slot
- Node-to-worker queue mapping
- Rule calendar and exception day translation
Dollar Universe session to DAG
Before
# Dollar Universe session (simplified)
SESSION: DAILY_ETL
RULE: MON-FRI 01:00
UPROC: EXTRACT
NODE: UNIX01
SCRIPT: /opt/etl/extract.sh
UPROC: LOAD
NODE: UNIX01
SCRIPT: /opt/etl/load.sh
DEPENDS: EXTRACTAfter (Airflow)
# Apache Airflow DAG
from airflow.decorators import dag
from airflow.operators.bash import BashOperator
from pendulum import datetime
@dag(schedule="0 1 * * 1-5", start_date=datetime(2025,1,1), catchup=False)
def daily_etl():
extract = BashOperator(task_id="extract", bash_command="/opt/etl/extract.sh")
load = BashOperator(task_id="load", bash_command="/opt/etl/load.sh")
extract >> load
daily_etl()Ready to migrate from ORSYP Dollar Universe?
Talk to a BatchFoundry engineer about your workload estate.
Talk to an Engineer