Cost-Efficient High-Scale Systems with Azure Cosmos DB | Patrick Oguaju | Azure Cosmos DB Conf 2026

Azure Cosmos DB gives developers incredible scale and flexibility — but without deliberate design, costs can spiral fast. Patrick Oguaju (Software Engineer at Next) walks through how his team re-architected a real production system to dramatically reduce Cosmos DB costs without sacrificing performance, availability, or developer velocity. The session covers: • How poor partitioning and indexing silently inflate RU consumption • Practical data-modeling patterns that cut read/write costs • Strategic use of TTL, Change Feed, and data tiering • When NOT to use containers, indexing, or multi-region writes • How observability exposed hidden cost leaks Demo-driven, grounded in real engineering decisions from high-volume, globally distributed systems. 👤 Connect with Patrick Oguaju 📝 Patrick C. Oguaju is a software engineer with hands-on experience designing and operating high-scale, cost-sensitive distributed systems on Azure. He has worked extensively with Azure Cosmos DB in production, supporting data-intensive workloads across various industries, including retail, content platforms, and cloud-native services. His work spans data modeling, migration, performance tuning, observability, and cost optimization. Patrick focuses on practical engineering lessons, including the trade-offs, failures, and insights that only emerge in real systems at scale, and enjoys sharing patterns that developers can immediately apply in their own architectures. 💼 https://www.linkedin.com/in/patrick-oguaju ✍️ https://medium.com/@pyrosv 🎤 https://developer.azurecosmosdb.com/conf/#speaker/patrick-oguaju --- Azure Cosmos DB Conf 2026 is a free virtual developer event focused on building modern, scalable applications with Azure Cosmos DB. Sessions feature Microsoft engineers, product teams, and members of the global developer community sharing real-world solutions and best practices. Originally streamed on April 28, 2026 | 9:00 AM – 2:00 PM PT 🔗 Useful Links • Explore all sessions o