Cerebro: Bringing together data scientists and BI users on a common analytics platform in the cloud

Royal Caribbean Cruise Lines operates in 47 countries with more than 60,000 employees and over 50 cruise ships, each equipped with floating data centers, which can support up to 7,000 guests at a time. The company’s systems generate large and valuable datasets that its data scientists and business analysts use to create new products, improve the operational efficiency of the fleet, and ensure the safety of its customers and employees, along with many other use cases.

Juan Bengochea offers an overview of Cerebro, a platform developed at Royal Caribbean Cruise Lines for analytics that serves the needs of both its data scientists and analysts from a common system. By combining the best in open source projects, including Hadoop, Parquet, Spark, and Apache Arrow, and cloud services like Azure Data Lake Store, Azure Databricks, and MongoDB Atlas, the platform helps Royal Caribbean’s data scientists and BI users operate more independently through a common infrastructure.

Juan outlines the architecture for Cerebro and explains how it was developed. Along the way, Juan describes the different needs of each data consumer function, as well as the various data privacy requirements given the global nature of the business, and how the platform addresses them and discusses future work.