Rocket^® Data Replicate and Sync

Rocket^® Data Replicate and Sync (RDRS) is a real time change data capture (CDC) and synchronization solution that keeps data continuously aligned across IBM^® mainframe, distributed, and cloud platforms. It enables organizations to modernize and operationalize mainframe data without disrupting production systems or relying on batch processes.

RDRS eliminates data latency, data drift, and operational risk created by batch extracts and fragmented replication tools. It enables trusted, real time access to mainframe data for cloud platforms, analytics, AI, and modern applications while the mainframe remains the system of record.

RDRS supports IBM z/OS^®, z/VSE^®, Linux on Z^®, Db2^®, and VSAM, as well as distributed and cloud targets including Kafka, relational databases such as MySQL^®, and major cloud platforms like AWS^® and Microsoft Azure^®.
 

Data replication copies data from a source system to one or more target systems, typically in one direction. Data synchronization goes further — it ensures that changes made in any connected system are propagated back to all others, keeping every environment consistent. RDRS supports both: one-way replication from mainframe sources to cloud targets, and full bidirectional synchronization where changes in either system are reflected everywhere. 

RDRS is purpose built for hybrid environments, streaming real time data changes from on premises mainframe systems to distributed and cloud platforms. This enables cloud adoption and modernization without forced migrations, data duplication, or re architecting core applications.
 

Yes. RDRS reads the Db2 for z/OS transaction log to capture inserts, updates, and deletes as they occur, without polling the database. This delivers sub-second data freshness to downstream targets with no impact on Db2 production query performance. 

Yes. RDRS supports bi directional synchronization across mainframe, distributed, and cloud environments. This enables coexistence scenarios where multiple systems can safely read and write data while remaining synchronized.
 

Yes. RDRS natively targets Apache Kafka, Confluent, AWS Amazon MSK, and other streaming platforms. This enables event-driven architectures where downstream consumers — including AI models, microservices, and analytics engines — receive mainframe data changes as real-time events rather than batch snapshots. 

Traditional batch replication and on-mainframe ETL consume significant MIPS because data transformation logic runs on mainframe compute. RDRS moves transformation processing off the mainframe to cloud compute — capturing data at the log level on the mainframe and applying transformations on the target side. Customers typically see meaningful reductions in mainframe CPU consumption for data integration workloads. 

RDRS respects existing mainframe security controls and works with encrypted data without weakening protections. Data integrity is maintained throughout replication and migration, enabling secure sharing of sensitive enterprise data across platforms.
 

RDRS captures every committed transaction at the source and applies changes in order on target systems, preserving transactional consistency. This prevents data drift and ensures downstream systems always operate on accurate, up to date data.
 

RDRS continuously streams trusted operational data into data lakes, data warehouses, and streaming platforms, ensuring AI models and analytics workloads run on current production data rather than stale batch snapshots.
 

Mainframe systems use EBCDIC encoding, packed decimal formats, and binary data types that require conversion before cloud platforms can consume them. RDRS includes an extensive built-in code catalog of codepage translations and data type conversions, along with support for custom transformation scripting (including REXX), to handle any mainframe data format without manual coding. 

RDRS pipelines are designed for operational resilience. In the event of a failure — network interruption, target system outage, or a planned maintenance window — pipelines are fully restartable from the last confirmed checkpoint. Built-in rollback and recovery controls ensure that no data is lost or applied out of order, and that replication resumes accurately without manual intervention,

RDRS pipelines are designed for operational resilience. In the event of a failure — network interruption, target system outage, or a planned maintenance window — pipelines are fully restartable from the last confirmed checkpoint. Built-in rollback and recovery controls ensure that no data is lost or applied out of order, and that replication resumes accurately without manual intervention,