ISV Provider - Data Replication Across Regions Using Azure Kubernetes Services

Client

Customer was a large ISV provider in the healthcare industry, with thousands of customers and employees spreading globally. They wanted a reliable solution for hosting their business-critical applications for high availability, security, and scalability.

Challenges

• The client had an existing set of applications developed using legacy architecture and wanted to move them to Azure Kubernetes Services (AKS).
• The AKS solution was not designed and tested by the client. Instead, they assumed that it would work seamlessly.
• The client implemented redundancy in a traditional monolithic application way through data replication.
• The mission-critical applications needed failover capabilities in the AKS environment, but the database systems were not designed to support the failover in the AKS environment.
• The client had hard dates to go live due to business priorities, and the dates are non-negotiable.

Solution

• Collaborated closely with Microsoft and database product teams to come up with a robust solution.
• Customized the configuration to create data replication across Kubernetes clusters to replicate work across regions.
• Wallero automated the scripts for internal monitoring of data replication. This allowed the failover to happen automatically under faulty conditions.
• Application traffic monitoring was configured to track the health of the Kubernetes environment. It automatically switched the traffic to the secondary Kubernetes clusters in case of the primary cluster failure.

Value

• Found the solution and configured the Azure Kubernetes platform to enable the migration for their mission-critical applications.
• Mission-critical services were running in a highly available environment meeting the uptime and business continuity requirements.
• Dramatically improved the scalability and reliability of the overall platform.
• Almost impossible timelines were met.
• We offered the overall solution and support at a very cost-effective model.