Zurich Insurance had been looking to adopt Kubernetes to help migrate and deploy applications to the cloud — modernizing their application infrastructure across the board. So when Elastic released the alpha version of Elastic Cloud on Kubernetes (ECK), the company jumped on the opportunity.
Brian Keegan's team ran a proof of concept to test the viability of ECK on Microsoft Azure Kubernetes Services (AKS), working with the Elastic Consulting and Microsoft teams to brainstorm on architecture and use cases, and to validate the approach with best practices. Elastic consultants supported during the development of a proof of concept, creating an ideal training foundation for the IT team at Zurich Insurance. The Elastic team collaborated with Zurich to establish ways to capture logs in Kubernetes and helped prepare everything for production, with recommendations on Azure components, instance sizes, and storage classes. The engagement provided a solid launching point for the implementation of ECK, which is secure by default, and supports declarative users and roles.
"Our experience with Elastic Consulting was excellent," says Brian Keegan. "Their guidance was instrumental in meeting an aggressive timeline. We completed the implementation of the minimum viable product up to UAT in less than six weeks, and that was our green light for going into production."
The shift to Elastic Cloud has dramatically simplified upgrades and infrastructure changes. That means Zurich Insurance gets an optimal return on investment in the Elastic Stack by eliminating many legacy solutions that hindered growth, revenue, and scaling, while freeing staff to focus on improving products and services instead of spending time upgrading and addressing timeouts.
"In the past, performing an upgrade required a huge project involving a lot of people from across teams — and it could take several weeks to complete," says Evgeny Sitnikov. "With the ease of use of Elastic, today we can make infrastructure changes in hours instead of days or weeks."
Evgeny Sitnikov and Brian Keegan expect Elastic usage to grow, with data flowing in from multiple sources. To manage the data cost-effectively, they're making full use of the cold, warm, and hot nodes in ECK — continually rotating older data onto less expensive cold storage while retaining the data for accounting, regulatory and other purposes.