Connect your AVEVA PI System with Azure, Google Cloud,
or your on-premises data platform.
Unlock real-time analytics and AI – without disrupting
your PI environment
Architecture overview: how PI Data Integrator works
The diagram below illustrates the end-to-end architecture of PI Data Integrator – from PI data ingestion through processing and orchestration to streaming or batch export into
cloud and on-premises data platform destinations.
What PI Data Integrator enables:
Real-Time and batch data ingestion
High-frequency data can be ingested using two modes, depending on the requirements of BI dashboards and monitoring workloads.
Historical backfilling
Full synchronization of PI Data Archive – essential for long-term analytics and AI.
AF structure support
ReadAF hierarchy, attributes and metadata.
Data quality & monitoring
Status tracking, retries, and error handling built in.
Modern data platform ready
Integration with modern data platforms in Azure, Google Cloud, and
on-premises environments.
User friendly interface (Smart RDM)
A simple UI for configuring, scheduling and tracking performance of data flows.
Stop fighting with PI data
This e-book explains why the AVEVA PI ecosystem, while essential for collecting industrial operational data, often becomes a bottleneck when organizations try to use it for analyticsand AI.
Key pain points the e-book addresses:
Analytics workloads putting performance pressure on the PI System, which is designed primarily for real-time data acquisition.
Slow and complex data access for BI and data science teams that depend on PI as their main data source.
Fragmented and duplicated data extraction pipelines built by different teams across the organization.
Limited scalability for advanced analytics and AI, especially when processing multi-year industrial datasets.
Data quality, structure, and reporting challenges that arise when raw
time-series data is used directly for business analytics.
Enable enterprise-wide access to PI data without affecting
performance of historian.
Accelerate analytics projects with streaming, ready-to-use data sets.
Support AI use cases with complete historical backfills.
Harmonize data across multiple sites for unified reporting.
Reduce manual engineering work by replacing scripts and CSV
exports with automated data streams.
Build a modern data architecture on top of the existing PI Installation.
Contact us
Ready to modernize your PI data? Let’s talk.
Whether you want real-time dashboards, full historical analytics or a cloud migration path – PI Data Integrator gives you a reliable, scalable and low-effort way to unlock the full value of your PI data.
Karol Woźniak
AI Data & Cloud Unit Manager
FAQ
What is PI Data Integrator?
PI Data Integrator is a reliable, scalable solution for transferring AVEVA PI System data (PI Data Archive + AF) into cloud platforms such as Azure, Google Cloud or on-prem data platforms.
It supports real-time streaming, batch processing, full historical backfilling, and includes a user-friendly management interface (Smart RDM).
Does PI Data Integrator affect PI System performance?
No. PI Data Integrator uses controlled, efficient data extraction mechanisms.
It supports both snapshot and archive modes without overloading PI, and can run parallel streams with monitored performance and retry logic.
Can PI Data Integrator run fully on-premise?
Yes. The solution supports on-premises as well as cloud-native deployment models.
Does it support multiple PI Servers or multi-site integrations?
Yes. PI Data Integrator canread from multiple PI environments and consolidate data into a unifiedcloudor on-prem data platform. It alsosupportsmulti-sitearchitectures with parallelstreams.
What types of data can PI Data Integrator process?
PI Data Integrator supports PI data types:
PI Data Archive
PI Asset Framework
Does PI Data Integrator support full historical backfilling?
PI Data Integrator can:
automatically or manually backfill PI Data Archive
recover missing batches
synchronize history for analytics and AI
Does PI Data Integrator support AF hierarchies?
Yes. PI Data Integrator supports reading the AF structure, attributes, metadata, and can reflect the hierarchy inside the target environment.
AF monitoring is also supported – changes are tracked and transferred.
Does it support both real-time and batch processing?
Yes.PI Data Integrator offers two ingestion modes:
Real-time streaming (near real-time flow of snapshot data into cloud endpoints)
Batch mode (scheduled, filtered extractions for analytics workloads)
How is data quality and reliability ensured?
PI Data Integrator provides Smart RDM UI for:
automatic retries and error handling
detailed status tracking
monitoring dashboards (Grafana)
logging and auditing
Does it require changes to the existing PI System?
No. PI Data Integrator works non-invasively, not requiring modifications to PI configuration, AF models or interfaces.
Is there a demo available?
Yes – you can schedule a live demo where our engineers will demonstrate the current features of PI Data Integrator.
Download case study
Enter your email address to download the case study