Enterprise data platform architecture evolution
Client: Comcast
Our client, a multinational $275 billion telecommunications conglomerate, had trusted Turnberry for years as a strategic partner in evolving its cloud adoption. The client turned to us again to advise its data architecture and cloud migration plans.
Client challenge
Our client has recently consolidated its major big data teams into a single group called DX Architecture. This group was responsible for managing an aging, expensive, and stress-inducing data platform primarily based on on-premise infrastructure, including Teradata, Kafka, and Hadoop. Relying on this platform for enterprise analytics and reporting was often a matter of luck, and gaining widespread adoption among the company’s numerous siloed teams proved impossible. To address these issues and improve its ability to fulfill enterprise-wide data requirements, the company embarked on a modernization journey, transitioning to an AWS-based, cloud-native platform to revolutionize enterprise data capabilities.
Our solution
A small, high-powered Turnberry team collaborated closely with DX architects and tech leads to gain a deep understanding of their aspirations, perceived use cases, and long-term vision for a feature-rich, consumer-friendly, and future-proof cloud-native enterprise data platform. The Turnberry team then established a validated reference architecture for the platform, comprising a diverse set of tools and solutions for data integration, ingestion, processing, storage, and querying strategies. Turnberry provided comprehensive guidance for each of these components individually and holistically as part of the entire platform. Additionally, we delivered detailed plans for migrating legacy products and their associated data to this next-generation solution, and later provided staffed engineering teams to support the ongoing efforts of this group.
Results
The platform has not only successfully come to life as designed but has also undergone further positive evolution. Hundreds of previously self-managed and siloed data teams have successfully onboarded into the platform, benefiting from robust and streamlined tools to store, catalog, and gain insights from their data. This integration has fostered data-sharing among teams, benefitting both individual teams and the organization as a whole. Presently, the platform hosts an impressive 14 PB of data (including 6 million objects), handling 500,000 queries daily, managing nearly 2,500 different event streams for system integration, and executing 5,000 ingestion jobs. The consumer portal, which serves as the front-end for the platform’s services, enjoys an average of 2,000 monthly users. The DX enterprise platform has now become the go-to solution for the company’s data integration and analytics needs.
