Training Agenda

Realtime BI

Traditional Business Intelligence runs on batch data — yesterday's numbers, last hour's aggregates. Realtime BI closes the gap: dashboards that reflect what is happening now, driven by streaming data pipelines from Kafka, CDC feeds, or event streams. This training covers the architecture and tooling for real-time analytics: streaming ingestion, OLAP engines built for low-latency queries, and BI dashboards connected to live data sources.

1 day On-site, remote, or hybrid Up to 20 participants German or English
What We Cover
Live dashboards and analytics on streaming data
Module 1

Realtime Analytics Architecture & Ingestion

  • Batch vs real-time BI: latency requirements, cost trade-offs, use cases
  • Lambda vs Kappa architecture for BI: pre-aggregation vs query-time computation
  • Streaming ingestion: Kafka → OLAP pipeline patterns
  • CDC as a BI data source: Debezium → Kafka → analytical store
  • Apache Druid: architecture (ingestion, query, coordination, historical nodes), rollup, pre-aggregation
  • ClickHouse: columnar storage, MergeTree engine, materialized views, real-time inserts
  • Apache Pinot: low-latency OLAP for user-facing analytics, startree index
  • Materialized views: maintaining pre-aggregated tables for sub-second queries
  • Time-series data: TimescaleDB for metric-like BI workloads
Module 2

BI Tooling & Dashboard Delivery

  • Grafana as a BI tool: connecting to ClickHouse, Druid, Pinot, and TimescaleDB
  • Apache Superset: connecting to real-time databases, chart types, dashboards
  • Metabase: self-service BI for business users on real-time data
  • Embedding dashboards: iframe embedding, token-based authentication, whitelabeling
  • Caching strategies: query result cache, time-bucket caching
  • SLA for real-time BI: freshness guarantees, monitoring pipeline lag
  • Data modeling for real-time: star schema vs wide tables for OLAP
  • User-facing analytics: multi-tenant real-time dashboards at scale
  • Cost model: ClickHouse vs Druid vs BigQuery BI Engine for real-time workloads
Learning Outcomes
What your team walks away with

Data engineers and architects who can design and build a real-time BI pipeline — from streaming data source through low-latency OLAP storage to live dashboards.

Book the Realtime BI training

Combines well with Apache Kafka, Apache Flink, or Big Data Engineering for a complete modern data platform training program.

Get in touch