On-prem to AWS migration for 100+ node cluster

A prominent west‑coast online gaming company was outgrowing its aging data center and data warehouse. AWS engaged our team (heritage at Full 360) to assess whether the workload could be migrated to the cloud.

The Challenge

A prominent west coast online gaming company was outgrowing their aging brick-and-mortar data center and data warehouse. As is so often the case with enterprise implementations, the need to manage upgrades, add computing resources, and ensure uptime were outside the client’s core competency. Further, the data warehouse quickly outgrew budget constraints.

The client worked with AWS to migrate their more‑than‑100‑node Vertica cluster to the cloud. Other firms said it couldn’t be done—no cluster that large had been built in the cloud. Confronted with this challenge, AWS and the client approached our team for expertise.

The Result

We proved it could be done. The team built a complete replica of the data cluster in AWS—migrating nearly a petabyte of data in the process.

However, it wasn’t merely a proof of concept; we demonstrated that such a cluster was reliable in production and that it would meet or exceed strict performance standards.

We built this massive Vertica cluster using deep expertise in imaging, DevOps, and service discovery. We demonstrated how to tear down and rebuild the entire infrastructure in an automated, predictable, and repeatable way.

Building the entire 100+ node cluster took only 48 hours, once the data warehouse backup become available in S3. This is incredibly fast for such a massive dataset, and clearly demonstrates the power of running a data warehouse in the cloud. Migrating a data warehouse to the cloud offers incredible flexibility: it can be moved geographically, destroyed and rebuilt, replicated, and upgraded. The cloud also removes the need for costly and time-intensive hardware maintenance.