Case Study: Expedia Group achieves faster, lower-cost cross-region data access with Alluxio

Unify Data Lakes Across Multiple Geographic Regions In The Cloud

Expedia Group, a global travel technology company, needed a better way to access petabyte-scale data spread across AWS regions. Its data platform teams were blocked by the high cost, slow speed, and operational complexity of manually replicating cross-region data into a central lake, creating long delays and error-prone workflows. To solve this, Expedia Group adopted Alluxio to federate its multi-region data lakes and support analytics and AI workloads across Spark, Trino, Hive, Databricks, and JupyterHub.

With Alluxio placed between S3 and the compute engines, Expedia Group unified cross-region data access without repeated replication and improved performance through caching. The result was faster data availability, simpler management, and an estimated ~50% reduction in S3 egress costs for frequently accessed tables. Expedia Group also said Alluxio improved manageability and performance enough that it plans to make it the default approach for cross-region data access across its main data lake clusters.

Expedia Group

Jian Li

Senior Software Engineer