Partnership
better together: databricks + eot.AI
If you want to maximize the value of your SCADA and Historian systems with your Databricks data lake—without added complexity—you’re in the right place. Discover how to unlock AI, Machine Learning, and Data Science to drive smarter industrial operations and boost productivity.


Databricks + EOT.AI: together for advanced operations
You’ve selected Databricks as your platform for hosting your industrial operational data (OT) from Scada and Historian systems, but implementing effective data pipelines that transfer historical and real-time data comes with challenges. While many options seem viable at first glance, they often involve intermediary cloud services or costly custom integrations. These approaches result in technical debt, lack scalability, are difficult to maintain, and worst of all, deliver “real-time” data in minutes instead of milliseconds.
But what if there’s a better way?
A solution trusted by some of the largest energy companies today, transferring billions of time-series OT records daily to Databricks. One that takes days, not weeks or months, to get fully scaled, deployed, and operational. One that gives you freedom by being open, reliable, secure, and faster than any other approach.
Twin Talk is the industrial OT Connector for Databricks
Twin Talk offers a simple, secure, and scalable way to transfer operational data from PI/AF, SCADA systems, historians, OPC-UA, MQTT, and other industrial sources into Databricks. This platform has changed the way industrial companies manage data, enabling cloud-based, event-driven, real-time architectures, from data lakes to mobile monitoring and surveillance apps, while unlocking valuable insights through analytics, AI, and machine learning. And it can do the same for you and your company.
Twin Talk bridges the gap between operational systems (OT) and cloud (IT) solutions. With low-latency, real-time data transport, Twin Talk enables a proactive, preventative approach to analyzing operational data. By applying modern data science techniques to the Databricks Data Intelligence Platform, companies can achieve predictive insights, reduced downtime, and operational efficiencies.
Real-time Monitoring with Databricks: Keeping a Finger on the Pulse
Twin Fusion’s integration with Databricks enables continuous monitoring and analysis of data from machinery and systems. This real-time observation allows manufacturers to spot inefficiencies as they occur, from minor operational glitches to significant system failures.
PROCESS OPTIMIZATION: MAKING THE GOOD GREAT
Twin Fusion and Databricks allow for the optimization of manufacturing processes. By analyzing data, the integrated system identifies areas where processes can be streamlined, leading to increased productivity and reduced waste.
EFFICIENCY AND COST REDUCTION: THE BOTTOM LINE
Perhaps the most tangible benefit of integrating Twin Fusion with Databricks is the impact on operational efficiency and cost. By leveraging real-time data and AI-driven insights, industrial companies can achieve significant cost savings.
How the integration works
1. SQL Warehouse API
Twin Talk uses the SQL Warehouse API to execute SQL queries to insert data in Delta Lake. It’s ideal for handling structured time series data and supporting SQL-driven ETL processes, ensuring a scalable and efficient data architecture.
- Insert Data: Use SQL INSERT to load data into Delta Lake tables.
- Manage Tables: Programmatically create and modify tables for seamless scalability.
2. Delta Sharing & Delta APIs
Twin Talk uses Delta Sharing for large datasets, to ensure efficient data ingestion and sharing, leveraging Delta Lake’s ACID transactions for consistency and scalability.
- Data Ingestion: Append, update, or delete records with Delta APIs, ensuring data integrity.
- Scalability: Ideal for large-scale, continuous data ingestion of AF and PI data.

Common applications built with Twin Talk and Databricks:
- Predictive Maintenance
- Resource Management and Process Optimization
- Production Efficiency and Quality Control
- Asset Performance and Production Optimization
Why customers use Twin Talk
Twin Talk, available across all three major clouds, delivers immediate impact for operators, data scientists, business analysts, and digital transformation executives. Embracing a simple, scalable, and secure data access strategy offers significant benefits, including:
-
Getting out of POC purgatory
Twin Talk reduces the time required to move from proof of concept (POC) to a functional system, shrinking deployment cycles to just a few days. By securely connecting your Databricks predictive analytics and operational efficiency applications to your data, you eliminate critical barriers. -
Scaling from pilot to production
With Twin Talk’s “no touch” externalization of identity and access management, scaling your solution from pilot to full production across geographic, logical, and organizational boundaries becomes seamless. -
Delivering value faster
Twin Talk allows AI experts and data scientists to securely test-drive more AI, ML, and analytics solutions, helping them find the best tools to deliver transformative value quickly. -
Making sense of data
Twin Talk understands operational data models and formats, translating and enriching them into formats easily consumed by cloud-based analytics, machine learning, and AI algorithms in real time.