databricks Databricks Blog ·

Indra Unifies EV Charging Data on Databricks, Enabling Self-Service Analytics

blogdatabricksarchitectautomotive
announcement

Indra Renewable Technologies consolidated its fragmented EV charging, fleet, and operational data onto Databricks, moving away from multiple Azure tools and duplicated pipelines. This platform consolidation reduced sprawl and overhead, establishing improved governance and a single source of truth. The shift enabled self-service analytics and automated reporting for business users, replacing manual processes and brittle functions. Ultimately, Indra achieved significant cost savings, faster query performance, and built a foundation for future streaming and AI initiatives.

  • Addressing Data Fragmentation with Databricks Consolidation
  • Streamlining Data Pipelines with a Medallion Architecture
  • Enabling Self-Service Analytics and Automated Reporting
  • Significant Cost Savings and Performance Improvements
Enhancements (2)
  • Streamlining Data Pipelines with a Medallion Architecture

    Indra replaced its complex, multi-stage fleet data pipeline, previously relying on multiple Azure functions, with a simplified medallion architecture on Databricks. This involved ingesting raw data into a bronze layer, transforming it in silver with PySpark, and publishing business-ready gold tables, resulting in one codebase, fewer moving parts, and 60-70% performance improvements.

  • Enabling Self-Service Analytics and Automated Reporting

    Manual KPI tracking and Power BI reporting were replaced by automated AI/BI dashboards powered by governed gold tables in Unity Catalog and served via a serverless SQL warehouse. Additionally, Genie Agents were implemented to provide conversational analysis over the data, allowing business teams to get answers in plain English without relying on the data team.

Notes (2)
  • Addressing Data Fragmentation with Databricks Consolidation

    Indra Renewable Technologies, a rapidly growing EV charging company, faced challenges with fragmented data across multiple Azure tools, duplicated pipelines, and escalating maintenance costs. The company opted to consolidate its diverse data estate onto Databricks to create a single, governed platform for analytics, reporting, and self-service access.

  • Significant Cost Savings and Performance Improvements

    The architectural shift delivered measurable benefits, including 80-90% business cost savings from the Synapse to Databricks migration and 90% storage cost savings. Query latency improved significantly from 24.4 seconds to 3.5 seconds, contributing to a reduced per-active-user cost.

Read the original announcement →

https://www.databricks.com/blog/how-indra-unified-ev-charging-data-databricks

Related releases