In this session you will learn about what it takes to process joins in distributed databases and then we’ll take a deep dive into how Apache Druid does joins in each of its processing engines. Druid’s native query engine is designed for fast queries and it can do joins, we’ll talk about query design to take advantage of how it works. You will also learn its limits and when a use case is better resolved by the Multi-Stage Query engine (MSQ). We’ll review the current state of MSQ-based joins and when to use broadcast vs sortMerge join algorithms based on how each works and how they use resources.
Physical Hardware, Digital Analytics: IoT Challenges, Best Practices, and Solutions
Electric vehicle maker Rivian and German startup Thing-it were kind enough to talk us through how real-time data and analytics play a key part in the evolving landscape of IoT (Internet of Things). The wealth...
Watch nowAnalyzing streaming data with Apache Druid
Streaming data is not only data in motion—it’s a potential source of valuable insights, ready to be harvested and utilized. The challenge is to analyze streaming data at scale and extract these insights—before...
Watch nowReal-Time Analytics in the Real World
Engineering teams increasingly have to deliver insights in real-time. But as they aim to reduce latency from event-to-insight, they also face the challenge of dealing with larger and more complex data and concurrent...
Watch now