Real-time Analytics for Data-Driven Applications
SpringOne Platform 2017
Milind Bhandarkar, Ampool
"To provide hyper-personalized digital experiences in the emerging market transformation, innovative enterprises are building modern data-driven applications to deliver continuing value to their always-connected customers. Such applications need to utilize closed-loop deep insights to influence their users' behaviors in real-time. However, the traditional ways of capturing users' interactions, transporting data to large data warehouses or data lakes, further away from applications, and processing these data across multiple slow stages cannot meet the real-time expectations of both customers and businesses.
What if one could capture, analyze, and serve data from a highly concurrent, high-performance data store powering these applications? In this talk, we'll present a memory-centric Active Data Store (ADS), powered by Apache Geode, to meet the exigent demands of modern applications while providing operational simplicity. Ampool's ADS allows fast ingest and storage of 'hot' app data, in situ updates and analysis, and data serving from the same scalable distributed in-memory data store. As the data cools (ages), Ampool ADS automatically tiers data to warm and cold secondary stores. By speeding analytics several-fold, Ampool enables feeding actionable insights back to applications, driving decisions in a closed loop.
We will demonstrate the applicability of Ampool ADS for such an app by serving all data-access patterns from a single memory-centric store."
