The CONNX Open Systems Event Replicator is a real-time data replication tool that propagates changes from a source database to a target database.  With the Open Systems Event Replicator, events that affect data in the source table, such as updates, inserts and deletes, will automatically be propagated to the target table in real time.  


The Open Systems Event Replicator has an easy-to-use graphical interface that deploys installation and configuration information to the source and target database servers.


The Business Case

Because it can store and process multi-value fields, handle large volumes of data, and manage high-speed transactions, Adabas is ideally suited to house the critical business data needs of today's fast-paced companies. With Adabas, applications are able to save and retrieve complex data structures in a single physical record and in a single operation.  As a multi-value database, Adabas has significant transactional performance advantages over relational databases, which would require several joins between many tables to accomplish the same task.


Although multi-value Adabas excels in transactional performance, relational databases excel in analytical capability. Relational databases also have a large selection of analytical and reporting tools available for companies to use. The Open Systems Event Replicator for Adabas give companies the best of both worlds: superior multi-value database performance and SQL-based tools for data reporting and analysis.


The Options

There are several ways to move Adabas data into a relational database:


  1. The extract and load approach


Most relational databases have "bulk mode" import tools that will read data from a text or binary file and place it into the database. A program is written to extract the data into a text file (using the same formats as the relational schema) and transfer the text files from the Adabas system to the relational database system. The relational database table schemas are manually created and a database utility is run to load the files into the table(s).


This was the common approach used in the past simply because there was no other way to accomplish the task.  But there are several problems with this option:

    1. There are many manual steps involved in the process.  

    2. Only a snapshot of the data is moved into the relational database. Each time current data is required in the relational database, the process must be repeated.


  1. Using the relational database vendors ETL tools


Many relational databases, such as Oracle, SQL Server, and DB2, have GUI-based tools that make it easy to move data from other sources into their database using an OLEDB or ODBC driver. SQL Server is an example of a relational database that comes with a set of powerful data management tools. Using SQL Server Integration Services (SSIS) combined with an OLEDB Driver for Adabas (Adabas ConnecX SQL Engine, it only takes a few clicks to move a snapshot of Adabas data into SQL Server.


These tools make the extraction and load of the data very easy in comparison to bulk loading data, but there are still problems:

    1. It's a manual process.

    2. It's a data snapshot. The process must be repeated every time current data is needed.


  1. The real-time changed data capture approach


Change data capture detects changes to data on a "source" database and replicates those changes to a "target" database in real-time (sub-second replication). The Open Systems Event Replicator can replicate data from one Adabas database to another Adabas database for backup or disaster recovery purposes. It can also replicate Adabas data to relational databases.


There are two advantages to this technology:

    1. "Fire and forget." Once the replication has been established, no further effort or manual steps are required. The replication will continue to run in the background until otherwise directed.

    2. The data in the Adabas target or the target relational database is not a one-time snapshot, but rather a living copy of the data that is kept in real-time synchronization with the source data in Adabas. This allows transactional applications to continue to benefit from Adabas high-speed performance and opens the door for using SQL-based data analysis and reporting tools against the relational copy.