Modernized configuration management to support growth from 2 to 15 clusters
“The most important thing for us is that all of our clusters look the same—have the same operators, the same configuration—whatever their purpose and wherever they are running,” said Karnutsch. “Red Hat Advanced Cluster Management ensures all are configured the same way whether developers decide to deploy them on premise or in the Azure cloud.”
Previously, Porsche Informatik took a ClickOps approach to configure management. This approach required developers to manually configure and deploy automated infrastructure across multiple clouds. While this may have worked for only 2 clusters, with 15 clusters and growing, ClickOps opens the door to inconsistencies and human error.
“Red Hat Advanced Cluster Management centralizes our configuration management and changes to cluster configurations can only be made through the policies stored in our GitLab environment,” said Karnutsch.
Importantly, Red Hat Advanced Cluster Management ensures Porsche Informatik can expand its OpenShift environment without expanding its operations team. The infrastructure-as-code—or rather configuration-as-code—approach makes cluster configuration and administration very simple. “Our operations team consists of 3 people,” said Karnutsch. “Without the advanced policy and configuration management, we would need a team of 5 or 6 people.”
The team can, for example, provide developers or the business with a new cluster—or even create a new cluster in the event of an issue—within a few hours. “This speed is only possible because the entire configuration is available as code with Red Hat Advanced Cluster Management policies,” said Karnutsch.
Ensured consistency and, thereby, compliance with ever stricter regulatory requirements
This consistency means Porsche Informatik can be confident that it is compliant with national and international regulatory requirements, including those set by the European Banking Authority (EBA) and the Digital Operational Resilience Act (DORA), across all its clusters. These requirements demand a higher-than-ever level of security and compliance across systems.
“Internal and external regulations are increasing, including configuration, audit, and backups,” said Karnutsch. “Red Hat Advanced Cluster Management has supported us with compliance from day one. We use its policies to create and deploy Compliance Operator, Kyverno policy engine, Kubernetes Network Policy, and Egress Firewall objects to meet regulatory requirements.”
Enabled an infrastructure-as-code approach while embedding DevOps in the operations team
DevOps has been introduced to Porsche Informatik’s operations team. “The infrastructure-as-code and configuration-as-code have enabled the incorporation of the DevOps methodology within our operations,” said Karnutsch. “Everything is now code. Everything is stored in a repository. Everything is repeatable.”
The team now runs its tests as pipelines and takes a GitOps approach to automating infrastructure provisioning and software deployment. Change requests are merged to avoid duplication of effort.
Engaging a Red Hat TAM brought extra reassurance to the operations team. “We can access professional help at any time,” said Karnutsch. “That makes us feel secure.” The TAM has helped the Porsche Informatik operations team internalize Red Hat expertise, including the Policy Generator team, the Red Hat Advanced Cluster Management product owner, and the GitOps team.
“Our TAM also helps us access other customers to understand how they are resolving challenges similar to the ones we face,” said Karnutsch. “They’re helping us connect with other companies working on establishing Egress policies.” A round table that takes place several times a year allows Red Hat customers to discuss their challenges and solutions with each other.