BlaBlaCar: Leveraging IoT for Enhanced Carpooling Services

• Analytics & Modeling - Machine Learning
• Platform as a Service (PaaS) - Application Development Platforms

• Cement
• Transportation

• Logistics & Transportation
• Procurement

• Construction Management
• Vehicle-to-Infrastructure

• Cloud Planning, Design & Implementation Services
• Data Science Services

BlaBlaCar, a leading long-distance carpooling platform, was facing challenges with its on-premises infrastructure. Initially built to support the French market, the infrastructure was struggling to keep up with the company's rapid global expansion. The company was experiencing issues with returning carpool search results consistently, processing payments, and connecting drivers and passengers. The engineering team was spending more time maintaining servers rather than focusing on enhancing platform features. The company needed a scalable, data-driven infrastructure that could support its growing global community and meet users’ needs in real time.

BlaBlaCar is a world-leading long-distance carpooling platform with a global community of 90 million drivers and passengers in 22 countries. The platform connects people looking to travel long distances with drivers heading the same way, allowing them to share the cost of the journey. The company aims to become the go-to marketplace for shared mobility, offering different modes of transportation including cars, e-scooters, and coach services between European cities. Since its launch in 2006, BlaBlaCar has saved an estimated €14 billion in traveling costs and operates a carbon-saving network, having saved an estimated 1.6 million tons of CO2 in 2018 alone.

To address these challenges, BlaBlaCar migrated to Google Cloud in 2018. The company adopted containerization, which allowed developers to create self-contained applications that could be deployed, tested, and updated independently. To automate deployment and manage these containers at scale, the company migrated to Google Kubernetes Engine (GKE), reducing operational overhead and simplifying daily operations. BlaBlaCar also leveraged BigQuery as their data warehouse of choice, enabling the data analytics and data science team to refine the platform’s passenger-driver matching algorithm. The company further adopted a service-oriented architecture with Cloud Bigtable, a scalable NoSQL database service for large analytical and operational workloads, and integrated these services with the fully managed relational database Cloud SQL.

• The migration to Google Cloud has enabled BlaBlaCar to focus on creating new features and improvements instead of managing infrastructure. The use of managed services has helped data scientists focus on matching the right passengers with drivers while Cloud Bigtable monitors database infrastructure. The company has also been able to automate more time-consuming tasks such as database provisioning and storage capacity management. This has allowed BlaBlaCar to leverage machine learning algorithms to help passengers and drivers efficiently find the best carpool at the right price. Furthermore, the company was able to quickly adapt to a fully remote working environment during the coronavirus pandemic and focus on technical migrations from its PHP monolith to a microservice-oriented architecture on Google Cloud.

• 99.9% availability on Google Cloud, providing a reliable form of transport for drivers and passengers

• Supports analysis of many petabytes of data per day with BigQuery

• App availability has gone up to 99.7% since the implementation, eliminating highly variable latency issues