Aravo Solutions Case Studies InMobi's Transition to Databricks Lakehouse: A Case Study on Streamlining Data Processing and Enhancing Advertising Effectiveness
Edit This Case Study Record
Aravo Solutions Logo

InMobi's Transition to Databricks Lakehouse: A Case Study on Streamlining Data Processing and Enhancing Advertising Effectiveness

Aravo Solutions
Functional Applications - Warehouse Management Systems (WMS)
Platform as a Service (PaaS) - Application Development Platforms
Cement
Telecommunications
Maintenance
Warehouse & Inventory Management
Construction Management
Picking, Sorting & Positioning
Cloud Planning, Design & Implementation Services
System Integration
InMobi, a company specializing in targeted mobile advertising, was grappling with the challenges of managing a complex legacy infrastructure and a multicloud data warehouse. The company's data processing requirements had escalated to 20+ terabytes per hour, leading to skyrocketing costs and the creation of data silos that hindered collaboration and data sharing. The proprietary nature of their multicloud data warehouse also posed significant challenges. InMobi's existing system was overly complex, prone to outages, and extremely costly to scale. The company realized that their current system was slowing down their ability to innovate and was keeping their engineering resources tied up in maintenance tasks. InMobi sought a single system that could address multiple issues, consolidate their disjointed systems into a single platform, and free up their engineers to focus on higher-value tasks such as developing machine learning and large language models.
Read More
InMobi is a global provider of enterprise platforms for marketers, specializing in mobile advertising. The company helps brands reach and engage consumers in a meaningful and cost-effective way by delivering targeted and personalized content, particularly on mobile devices. InMobi uses real-time customer data to drive engagement and deliver relevant ads. However, as the company's data processing requirements increased, they faced challenges with their existing multicloud data warehouse system. The company sought a solution that would streamline operations, reduce costs, and improve collaboration and data sharing. InMobi's goal was to free up their engineers to focus on higher-value tasks and improve operational agility and efficiency.
Read More
InMobi decided to migrate from their multicloud data warehouse to Databricks Lakehouse to unify their data warehousing, AI, and analytics workloads on a single platform. The company partnered with Databricks and Celebal Technologies to plan and execute the migration process. Despite the complexity of the migration, which involved over a decade of customizations, over 1 petabyte of data, 150 pipelines, and eight teams, the transition was successful. With the new lakehouse architecture, InMobi was able to fully leverage their robust customer data to deliver smarter, more personalized mobile advertising. They used Databricks notebooks for ad hoc analysis, Power BI for visualizations on top of Databricks SQL, and MLflow to build their next-generation AI platform. They also used Delta Live Tables for anomaly detection and Unity Catalog to govern access at the table and column levels, ensuring complete data lineage.
Read More
The migration to Databricks Lakehouse resulted in a unified, streamlined architecture that allowed InMobi to take full advantage of their robust customer data. The company was able to deliver smarter, more personalized mobile advertising and improve collaboration and data sharing. The new system also eliminated data silos and improved data discoverability. With the help of Unity Catalog, InMobi was able to govern access at the table and column levels and ensure complete data lineage. The team also experienced better reliability with more stable systems and a positive reputation boost with their customers. The migration freed up InMobi's engineers to focus on innovating in the mobile advertising space and deliver real-time personalization that drives value for both InMobi’s customers and their internal end users.
Infrastructure costs were reduced by 34% after the migration
Query speeds improved by 15%
Job failures decreased by 20% compared to their previous data environment
Download PDF Version
test test