5:30-6:30 - Presentation
6:30-7:00 - Questions
Digital transformation initiatives have unlocked large and fast-moving data sets including clickstreams, network telemetry, application monitoring and IoT devices. Analytics architectures have not kept pace, with most data still being run through existing “cold analytics” systems and tools designed for smaller and less time-sensitive workloads. “Hot analytics” denotes workloads where the responsiveness of the system is instantaneous and can support self-service data exploration, and where the data is extremely fresh, allowing for more informed decision-making.
The breadth of analytical systems in the world today demands a clear approach to selecting the right one for a given workload. In this talk we’ll discuss a temperature-based way of thinking, where workloads get “hotter” as they become more interactive, more concurrent, and more likely to need up-to-the-second data.
Apache Druid is a modern cloud-native, stream-native, analytics database designed for workflows where fast queries and instant ingest are important. Druid excels at instant data visibility, ad-hoc queries, operational analytics, and handling high concurrency. It is a strong candidate for being the workhorse system for hot analytics.
Speaker bio: Rachel Pedreschi is the VP of Community at Imply. A "Data Geek-ette”, Rachel is no stranger to the world of high-performance databases and data warehouses. She is a Vertica, Informix and Redbrick certified DBA on top of her work with Cassandra and has 20+ years of business intelligence and ETL tool experience. Rachel has an MBA from San Francisco State University and a BA in Mathematics from University of California, Santa Cruz.