processes orchestrated via process control with Camunda
records processed automatically per run
end users with transparent error analysis
migrated from on-prem Kubernetes
GEMA
Industry: Media
Headquarters: Berlin, Munich
Board Member, cimt Group
GEMA manages music rights for more than 100,000 members and over two million international rights holders. Large volumes of data are processed between different systems in complex, cross-team processes.
Before the project, error analysis was time-consuming because several teams had to check their data separately. Together with cimt, GEMA therefore implemented central process control with Camunda that orchestrates millions of records automatically and makes every process step transparent and traceable. The result is a scalable platform for cross-team processes – and the foundation for a modern, cloud-based data architecture.
GEMA processes large volumes of data from various specialist systems to correctly bill and manage the use of musical works. Several teams work together on complex, cross-system processes – yet coordination was largely manual. Data flows were documented in a decentralised way and held in different systems and formats.
When an error occurred, for example in transactions, a laborious root-cause analysis began: every team involved had to check its data separately. As data volumes grew – a single run could produce several million records – this became increasingly difficult. At the same time, the migration of the infrastructure to Google Cloud was coming up.
Together with GEMA, cimt developed central process control based on the Camunda 7 workflow engine. The project started with a proof of concept by a two-person team – a senior and a junior developer focusing on Java and Camunda. Four architecture principles support the platform:
System functions run as independent workers – similar to microservices. This makes the platform easy to extend and scale.
For every system, they describe messaging (JMS / Pub/Sub) and data structures – the stable basis for collaboration between teams.
Initially each system had its own worker. Unifying large parts of the logic made the architecture much easier to maintain.
Migration from on-prem Kubernetes: Cloud SQL for operations, Pub/Sub for messaging, BigQuery for archiving and tracking.
Operational data continues to be processed with high performance in the cloud, while historical data is replicated to BigQuery in parallel. This allows even large data volumes to be analysed efficiently over the long term.
Thanks to central process control, we can now trace exactly how data flows through our systems. That makes error analysis considerably easier.
Karolina Koric, Product Owner, GEMA
As the scope grew, so did the team. An initial task-board structure evolved into an agile working model with two-week sprints and a Scrum Master. Four practices ensure the quality and stability of the platform:
In every sprint, one team member is responsible for support and bug fixing.
Business acceptance of every ticket together with the business department.
End-to-end tests with Cypress and GitHub Actions.
Worst-case scenarios defined at an early stage.
In almost three years, more than 580 processes have run through central process control with Camunda. For the first time, the platform provides full transparency over data flows between the systems and teams involved.
When an error occurs today, it can be traced directly in the process flow: every processing step is documented automatically. Where several teams previously had to check their data separately, the cause now often becomes visible immediately in the orchestrated process – benefiting more than 100 end users.
Within GEMA, the solution has become a central part of the IT landscape and also serves as a pilot platform for new features and infrastructure changes. It also brings important governance benefits:
Working with cimt was a true partnership – the team quickly understood our processes and built a scalable solution.
Karolina Koric, Product Owner, GEMA
Central process control is being developed continuously and has sparked an important discussion about data quality and data management: in future, process control and data management will be more closely interlinked.
The project has also led to a follow-up initiative that modernises the underlying data architecture – with a more flexible structure that bundles references from different data sources and documents changes over time in a traceable way.
Learn more about our approach on our Camunda Platinum Partner page.
Many companies run business-critical processes on Camunda 7 and are asking when and how to switch to Camunda 8. As a Camunda Platinum Partner, we support you from the assessment of your process applications to the migration – or when starting new projects directly on Camunda 8.
Panasonic Europe digitalises hundreds of approval processes with one generic BPMN model and DMN decision tables in Camunda; cimt is responsible for the methodological design and supports the implementation.
For Berliner Wasserbetriebe, cimt migrated construction project control including long-running process instances to Camunda BPM and gradually replaced the Java EE monolith with microservices.
For GVL, cimt automated data analysis and cleansing with Camunda as part of a new billing procedure and implemented an extensive rule set with BPMN and DMN.
You are currently viewing a placeholder content from Turnstile. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
More InformationYou are currently viewing a placeholder content from Facebook. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
More Information
