by Peter M. Horbach

Mainframe data are often an obstacle to the use of new technologies

There are many reasons for migrating mainframe data or applications. This has already been discussed in another blog post.
Mainframe data are different - they have grown historically and are not compatible with the data storage systems under Linux, Unix, MS Windows and cloud platforms.

The BOS tcVISION solution is an all-encompassing solution for synchronizing mainframe data in real time with databases under Linux, Unix or Windows.

Transformations of the data are carried out automatically. The processing and transformation rules are stored in a repository.
A bidirectional replication and synchronization of the data is also fully supported with tcVISION. Intelligent loopback procedures ensure that in the case of bidirectional replication, data that has already been processed and committed is filtered out.
However, what should be done if mainframe data is available on backup or recovery media and this still has to be processed with data from the open systems

These data include, for example:

  • DB2 z / OS image copy
  • DB2 z / OS archive log files
  • IMS unload files (z / OS)
  • IMS log files (z / OS)
  • Sequential or VSAM files (z / OS and z / VSE)
  • ADABAS unload files (z / OS, z / VSE)
  • ADABAS PLOG files (z / OS, z / VSE)
  • CA DATACOM unload extract / backup files (z / OS, z / VSE)
  • CA DATACOM SPILL files (z / OS and z / VSE)

BOS offers a solution for all of these input sources without the need to install a software component on the mainframe.
The solution is: zSTREAM - transformation of IBM mainframe data for use for Big Data, Analytics, BI, Data hubs on non-mainframe platforms on premise and on Cloud platforms.

zSTREAM supports the methods and procedures BULK transfer and processing of change data which are known from tcVISION, but works exclusively under Linux, Unix or Windows as a JAVA application.
Similar to tcVISION, zSTREAM is also based on a repository which contains all information necessary for processing (metadata, transformation rules, processing rules, etc.).

The processing methods include in detail:

BULK transfer

Bulk transfer is an efficient transfer of entire databases and is suitable for

  • Db2 zOS image copy files
  • IMS unload files
  • Sequential files (and also VSAM unloads)
  • ADABAS unload files
  • CA-DATACOM unload and extraction files

These files are created on the mainframe and transferred to a system under Windows, Unix or Linux by means of a file transfer in binary form.
The bulk transfer is ideal for an "initial load" of a database or a periodic mass data transfer. All parts of a data exchange, extracting, transforming and entering the data into the target system, are carried out in a single step.

Processing of change data

Change data is processed from the following databases:

  • DB2 z / OS Archive Log files
  • IMS log files (z / OS)
  • ADABAS PLOG files (z / OS, z / VSE)
  • CA DATACOM SPILL files (z / OS and z / VSE)

As with the previously described BULK Transfer process, these files are created on the mainframe and transferred in binary and compressed form from the mainframe to a Linux, Unix Windows platform on premise or a cloud system. Processing with zSTREAM takes place there.

The productive data source on the mainframe remains unaffected.

The transformed data can be entered directly into any data target (e.g. SQL Server, Oracle, Db2 LUW, etc.) or transferred to a streaming platform such as Kafka.

Thus, zSTREAM is the suitable means to carry out transformations of IBM mainframe data for use for Big Data, Analytics, BI, Data hubs on non-mainframe platforms on premise and on Cloud platforms.

Peter M. Horbach has been active in the area of data synchronization and replication with more than 40 years of IT experience. He manages the international partner business for BOS Software and writes for our blog.

back to overview

zurück zur Übersicht