Treehouse Software Customers are Looking Upwards to Mainframe-to-Cloud Data Replication

The search is on for a mature, easy-to-implement Extract Transform and Load (ETL) solution for migrating mission critical data to the cloud.

Mainframe_to_Cloud

Treehouse Software’s tcVISION supports a vast array of integration scenarios throughout the enterprise, providing easy and fast data migration for mainframe application modernization projects, and enabling data replication between mainframe, Linux, Unix and Windows platforms. This innovative technology offers comprehensive abilities to identify and capture changes occurring in mainframe and relational databases, then publish the required information to an impressive variety of targets, both on-premise and cloud-based.

Mainframe-to-Cloud Case Use Example…

BAWAG P.S.K. is one of the largest banks in Austria, with more than 1.6 million private and business customers and is a well-known brand in the country. Their business strategy is oriented towards low risk and high efficiency.

BAWAG was looking to reduce the load on their IBM mainframe and as a result, reduce costs. The project involved offloading data from their core database system to a less expensive system, in real-time, and to provide read access from that system to the new infrastructure. The primary motivator for this data migration was the constantly increasing CPU costs on the mainframe caused by the growing transaction load of online banking, mobile banking, and the use of self service devices.

BAWAG ultimately migrated their online banking application to the cloud using tcVISION. Realtime Event-handling, Realtime Analytics, Realtime Fraud Prevention are only a few of the use cases that the bank’s solution currently covers.

BAWAG_Diagram

The bank decided to use tcVISION to migrate z/OS DB2 data into a Hadoop data lake (a storage repository that holds raw data in its native format). 20 Million transactions were made within 15 minutes.

Cost Reductions Seen Immediately

BAWAG is now seeing a 35-40 percent reduction of the MIPS consumption for online processing during business hours. After hours, consumption is less, because it is mainly batch processing on the mainframe. Currently, a volume of approximately 30 GB changed data (uncompressed) is replicated from DB2 per day.

In addition to the primary usage scenario, BAWAG can also cover additional use cases. This includes real-time-event handling and stream processing, analytics based upon real-time data as well as the possibility to report and analyze structured and unstructured data with excellent performance. The system can be inexpensively operated on Commodity Hardware and has no scalability limitations. Compared to the savings, the costs of replication (CPU consumption) of tcVISION are now very low.

Additionally, BAWAG plans to extend the use of tcVISION in the future, including implementation of real-time replication from ORACLE into the data lake.


Find out more about tcVISION — Enterprise ETL and Real-Time Data Replication Through Change Data Capture

tcVISION supports a vast array of integration scenarios throughout the enterprise, providing easy and fast data migration for mainframe application modernization projects and enabling bi-directional data replication between mainframe, Linux, Unix and Windows platforms. This innovative technology offers comprehensive abilities to identify and capture changes occurring in mainframe and relational databases, then publish the required information to an impressive variety of targets, both on-premise and Cloud-based.

_0_tcvision_connection_overview

tcVISION acquires data in bulk or via change data capture methods, including in real time, from virtually any IBM mainframe data source (Software AG Adabas, IBM DB2, IBM VSAM, IBM IMS/DB, CA IDMS, CA Datacom, even sequential files), and transform and deliver to virtually any target. In addition, the same product can extract and replicate data from a variety of non-mainframe sources, including Adabas LUW, Oracle Database, Microsoft SQL Server, IBM DB2 LUW and DB2 BLU, IBM Informix and PostgreSQL.


__TSI_LOGO

Visit the Treehouse Software website for more information on tcVISION, or contact us to discuss your needs.

tRelational / DPS Adabas-to-Oracle Success in South Africa

by Hans-Peter Will, Senior Technical Representative and Joseph Brady, Manager of Marketing and Technical Documentation at Treehouse Software, Inc.

Recently, Hans-Peter Will, Senior Technical Representative for Treehouse Software, traveled to South Africa to assist our partner Bateleur Software (pty) Ltd. with setting up a large public-sector customer’s data replication implementation using our Adabas-to-RDBMS tool set, tRelational / DPS (Data Propagation System).

Arriving in Johannesburg, Peter met with representatives from Bateleur and the IT Organization, where the key players discussed how tRelational / DPS was going to be used in the project. The customer initially wanted to populate sample data into Oracle, so Peter configured tRelational / DPS to process one of the smaller Adabas files to generate some data. He also recommended running an analysis with tRelational to determine whether the file contained a unique key. Peter took this opportunity to show the customer what other benefits they could realize out of the analysis information. Interestingly, they used a personnel file for analysis, and Peter was immediately able to show that 23 of the records had no gender entry and 180 of records had no surname. The customer was very pleased to see these revelations in the first analysis, and looked forward to identifying other data quality issues before commencing data replication.

Customer Replication Scenario with Treehouse Software Product Set…

_0_tReDPS_Replication_Scenario

The next step was to build the target structure in accordance with the Oracle DBA’s requirements. The DBA had specified that all columns were to be defined as VARCHAR2, except the date information. After the first model was completed, DDL and DPS parameters were generated and a quick materialization of data accomplished the desired result.

At the subsequent kickoff meeting, Peter provided a complete tRelational / DPS overview and discussed the target structure with the attendees and the Oracle DBA. The rest of the day was spent doing Adabas file implementations, analysis and modeling.

Setup was then completed for transferring extracted and transformed Adabas data into the customer’s Windows environment. Adabas Vista is used, so that one logical Adabas file was actually split into two files stored on different Adabas databases, and the customer wanted to combine them into the same target table in Oracle. While there was no unique descriptor, it was discovered that three fields in combination would make a unique key, enabling the model to be created to combine data from the separate physical files into a single Oracle table.

The team proceeded with file implementation, modeling, mapping, DPS executions, and resolving data issues. Various issues that were encountered, like invalid tab characters within the data, negative personnel numbers, duplicates in unique keys (maintained by the application) and the need to add an extra column to the output. These issues were resolved quickly by the customer’s staff.

Within a day, all the files were materialized and the PLOG copy process was modified so that from that point forward, every PLOG copy would automatically be processed through DPS Propagation to update the RDBMS on the target Windows machine.

The next day, Peter was asked by the customer, “How many of the files have been processed so far?”. Peter was pleased to report that every file was processed and was propagating successfully. The happy customer remarked that they never had a project that was completed this far ahead of the deadline.

Throughout the project, Peter never personally laid hands on a customer keyboard, but instead sat with staff, effectively training them and handing over comprehensive knowledge of tRelational / DPS. The customer was very excited to learn that their personnel can now easily use the product set to do any remaining work on their own.

A few days later, we received an e-mail from Bateleur:

“I had a very pleasant meeting with the customer today. They used tRelational to reject the non-unique keys, reran the Materialization, and reran DPS plus update into Oracle. The month-end update of Oracle that was taking nearly three days to complete, now takes five Minutes! Everyone delighted!”

Bjørn (Sam) Selmer-Olsen, Managing Director, Bateleur Software (pty) Ltd


About Treehouse Software’s tRelational / DPS Product Set

tReDPS_DIAGRAM

tRelational / DPS is a robust product set that provides modeling and data transfer of legacy Adabas data into modern RDBMS-based platforms for Internet/Intranet/Business Intelligence applications. Treehouse Software designed these products to meet the demands of large, complex environments requiring product maturity, productivity, feature-richness, efficiency and high performance.

The tRelational component provides complete analysis, modeling and mapping of Adabas files and data elements to the target RDBMS tables and columns. DPS (Data Propagation System) performs Extract, Transformation, and Load (ETL) functions for the initial bulk RDBMS load and incremental Change Data Capture (CDC) batch processing to synchronize Adabas updates with the target RDBMS.

Visit the Treehouse Software website for more information on tRelational / DPS, or contact us to discuss your needs.