How PERI Replaced Its Legacy Replication Stack — Connecting 55 Databases to Two Destinations Without Missing a Beat

Manufacturing

CDC from 55 Informix instances to two destinations simultaneously

The company needed to extract change data from 55 country-level Informix databases and load it into both SAP HANA and Azure Data Lake Storage Gen2 in Parquet format. CData Sync handles both destinations from a single replication layer, with minimal load on the source ERP systems.

Schema updates handled without manual rework per database

Source schema changes previously required 10 to 15 minutes of manual adjustment per database. With CData Sync's API and the team's DevOps automation, schema updates across all 55 instances are pushed programmatically, reducing a process that scaled with the number of country databases to one that does not.

SQL-based transformations that any reporting engineer can maintain

Where the competing tool used a proprietary transformation language that the team found complex and unfamiliar, CData Sync supports transformations in standard SQL. For a team already working in SQL for reporting, this removed a learning curve and made the pipeline logic maintainable without specialist knowledge.

When its CDC replication tool was discontinued in its existing form, PERI evaluated the market and selected CData Sync — connecting 55 Informix databases to SAP HANA and Azure in 110 jobs.

PERI SE is a global manufacturer of scaffolding and formwork systems, headquartered in Weißenhorn, Germany, with operations across multiple continents. Its Group IT team manages the data infrastructure that connects a proprietary ERP system, built on IBM Informix to two reporting environments: a legacy SAP BW on HANA layer serving Crystal Reports, Lumira dashboards, and a customer portal, and a modern Azure-based Delta Lake orchestrated through Databricks, feeding Power BI and Microsoft Fabric.

Because PERI's ERP is country-specific by design, there is a separate Informix database instance for each country the company operates in. At the time of the migration, that meant 55 source instances, all running the same schema, all needing to be replicated into both destinations. The team’s CDC replication had been running on the same platform since the ERP was first deployed. The tool had over two decades of history in the SAP and ERP integration space and had quietly become load-bearing infrastructure.

In early 2026, the vendor announced it was winding down that product line as part of a broader portfolio consolidation. The version PERI had been running would be discontinued. No migration path came with the announcement. The team decided to move off the platform and start evaluating alternatives.

The challenge: Replacing a deeply embedded CDC tool across 55 database instances

The core technical requirement was specific: CDC from Informix into SAP HANA. Change Data Capture was essential because running full extracts against the operational ERP databases was not viable. The team needed incremental replication that put minimal load on source systems. That constraint significantly narrowed the field of alternatives.

“CDC support for Informix and native destinations for both SAP HANA and ADLS Gen2 were the main criteria. We needed a way to extract only the changes without putting load on the ERP, and it had to land in the right place.”
Andreas Ott
Software Architect, Group IT — Information Architecture, PERI SE

PERI evaluated three options: the incumbent vendor’s updated product, a competing tool, and CData Sync. The first was ruled out on cost. The second was a serious contender, but it didn’t yet support SAP HANA as a destination at the time of evaluation. That capability arrived in April 2026, after the team had already completed its proof of concept with CData. CDC support for Informix, along with native destinations for SAP HANA and ADLS Gen2, were among the main selection criteria PERI couldn't compromise on. CData was the only option that met all three from the start.

Beyond the connector requirements, the team needed a tool whose transformation logic would be maintainable by the people who actually worked with the data. The competing tool used user-defined functions in a proprietary language the team found opaque. CData Sync's support for standard SQL queries in transformations was a practical differentiator for a team already fluent in SQL for reporting work.

Informix to CData Sync Architecture

The solution: CData Sync as the replication layer for a 55-instance ERP

PERI's team ran a proof of concept with CData Sync before committing, validating that CDC from Informix worked as expected and that the HANA destination performed reliably. The PoC confirmed both. With the evaluation complete, PERI moved forward with Sync across the full environment.

The architecture uses 110 Sync jobs in total, two jobs per country, one targeting SAP HANA for the data warehouse layer, and one writing Parquet files to Azure Data Lake Storage Gen2 for the Databricks pipeline. Each job contains between 150 and 160 tasks, corresponding to the tables being replicated. The Informix CDC engine handles incremental extraction, ensuring only changes are pulled from each source database rather than full refreshes.

PERI's team set up the jobs programmatically using CData Sync's API, which made it practical to configure 110 jobs across 55 instances without doing so manually. For ongoing schema changes, which happen periodically across the source databases, PERI's team built a DevOps automation that will push schema updates through a PowerShell pipeline, eliminating most of the manual work that previously took 10 to 15 minutes per database instance.

Transformations are handled in SQL within Sync, including custom columns, rounding logic, and trim operations keeping the pipeline logic in a language PERI's reporting team already knows and can maintain.

“What was decisive for us was the SQL query support for transformations. Our team works in SQL for reporting. The competing tool used its own language that nobody knew — SQL is something everyone understands.”
Andreas Ott
Software Architect, Group IT — Information Architecture, PERI SE

The outcome: 110 jobs running with schema automation

With 110 jobs now running across all 55 country instances, PERI's replication architecture, Informix CDC to SAP HANA and ADLS Gen2 (using Delta Parquet format) is fully validated and in production. The entire job setup is deployed programmatically via the Sync API, giving the team a repeatable, scalable approach that their previous tool couldn't match. Schema updates across all instances roll out automatically through the DevOps pipeline, with almost no manual effort required.

“We’re seeing far fewer errors and the integration is more stable. And the alerting is really reliable now. Before, when there were errors, sometimes we did not get a notification at all. With CData Sync, we actually know when something needs attention."
Andreas Ott
Software Architect, Group IT — Information Architecture, PERI SE

The result is that teams across operations, sales, engineering, and finance now have access to near-real-time operational reporting. CData Sync serves as the replication layer underpinning a broader analytics and AI modernization at PERI, as the team works to replace legacy SAP BW and HANA exports with data lake data as the foundation for Power BI reporting. This initiative is part of a larger transformation documented in a published Microsoft case study.

CData Sync: CDC replication built for complex, multi-instance environments

CData Sync gives data engineering teams a direct path from source to destination, including CDC from legacy databases like Informix without locking them into proprietary tooling or opaque transformation languages. For PERI, that meant finding a solution that met every technical requirement from day one: CDC from 55 Informix instances, dual destinations in SAP HANA and Azure, SQL-based transformations.

See how CData Sync fits your replication architecture

Get started today