************************************* *** IBM DataPropagator Relational *** ************************************* Data on the go for businesses on the move WHAT IS IBM DATAPROPAGATOR RELATIONAL? IBM* DataPropagator* Relational, the core product in IBM's Data Replication solution, is designed to meet your relational data copy needs for both operational and informational applications. DataPropagator Relational replicates data between databases in the DB2* family. Combining versatility with ease of use, it lets you specify copy intervals, enhance data for improved analysis, maintain historical change information, and control the impact of copy operations on system resources while delivering data across your enterprise. WHAT YOU CAN DO WITH DATAPROPAGATOR RELATIONAL Reengineer business processes. With DataPropagator Relational, you can replicate production data to servers across your company, and improve availability and responsiveness by moving data and applications to the point of each business transaction. Typically, this type of replication involves data subsetting -- for example, by region or branch. Unlike many replication products that support data subsetting only according to information contained in the replicated data, DataPropagator Relational supports data subsetting based on join predicates or subselects, allowing you to distribute data efficiently from normalized databases. Go mobile. DataPropagator Relational supports mobile computing where connections are infrequent and unpredictable. Key to this is the ability to initiate all data transfers from the mobile unit, whether downloading from a central server or uploading for consolidated processing. Improve decision-making effectiveness. DataPropagator Relational enables you to tailor data for maximum usability. Its unique relational staging table design makes available the power and flexibility of the relational database engine, allowing you to automate data enhancement as the data is copied to its target table. For example, you can: * Derive data using arithmetic, Boolean, or any valid SQL expression. * Aggregate data to produce sums or averages using SQL column functions. * Convert data by translating encoded fields to descriptive fields. * Consolidate data through joins or unions. * Generate histories to support trend analysis. Manage replication with your existing staff. DataPropagator Relational's graphical user interface (GUI) makes it easy to define what data, to where, and how often it should be copied. DataPropagator Relational automatically creates and loads copy tables. You can add, delete, or modify copy definitions while the rest of the system continues to run. Best of all, you can define your replication environment without changing, recompiling, rebinding, or relinking any applications. Minimize impact on production systems and networks. DataPropagator Relational is designed to efficiently use system and network resources. It employs a log-based change capture technique that minimizes impact on transaction performance, thus avoiding contention with base tables and inline transaction processing. It features optimizations to support various networked environments. You can specify distribution timing on a copy-by-copy basis, to minimize use of peak network periods or to take advantage of economy network prices. And because data reduction can occur prior to data transfer, you can avoid replicating hot spot activity across the network. Support mixed database environments. DataPropagator Relational expands the range of solutions available to you by supporting an open architecture. In particular: * DataPropagator Relational is built on standard SQL to leverage the database engine functions for data enhancement, network connectivity, and data security. * The data staging area is designed to support interoperability among multi-vendor sources and targets, between relational and non-relational formats, and among products from independent software vendors. * DataPropagator Relational replication directly supports multi-vendor sources and targets through DataJoiner*, IBM's multi-database server product. DATAPROPAGATOR RELATIONAL AND IBM'S DATA REPLICATION SOLUTION DataPropagator Relational establishes the base architecture for IBM's Data Replication solution. The IBM replication architecture is based on individual components that interoperate, letting you customize your data replication environment. Capture components capture changes as they occur in the source and store them in the staging area. Apply components read the staging area and apply those changes to targets or copy data directly from the source in full refresh mode. Administration components provide the GUIs for defining copy requests. DataRefresher and DataPropagator NonRelational allow for replication of non-relational legacy data. They can populate the data staging area with IMS* or VSAM data. DataPropagator Relational then enhances, distributes, and applies data to the copy tables to provide an end-to-end copy solution from legacy sources to client/server targets. NEW FEATURES IN DATAPROPAGATOR RELATIONAL RELEASE 2 DataPropagator Relational Release 2 expands on Release 1 functions as follows: * New Capture components support DB2 databases on AIX* and OS/2*. * New Apply components support DB2 databases on HP-UX and the Solaris Operating Environment. * Support for multi-vendor targets through interoperability with IBM DataJoiner. * Enhanced usability with the GUI: - source key columns are displayed in the subscription notebook - the source table name is set as the default target table name - users can specify a database in which to create the control tables - control tables are dropped when all subscriptions are canceled. * Improved performance and tuning: - with the new NOPRUNE parameter and the PRUNE command for Capture, users can defer pruning to non-peak periods which for some environments can mean up to 90 percent CPU reduction for the Capture program and 50 percent CPU reduction for the Apply program - you have the ability to invoke IBM or third-party utilities to initialize an AIX, OS/2, HP-UX, or Solaris copy as an alternative to SQL refresh. For more information, contact your IBM authorized software reseller or IBM marketing representative. (C) International Business Machines Corporation 1995 IBM United States Santa Teresa Laboratory P.O. Box 49023 San Jose, CA 95161-9023 5-95 All Rights Reserved References in this publication to IBM products or services do not imply that IBM intends to make them available outside the United States. * IBM, DB2, OS/2, AIX, DataHub, DataJoiner, DataPropagator, DataRefresher and IMS are trademarks or registered trademarks of International Business Machines Corporation. All other products are trademarks or registered trademarks of their respective companies. GC26-8463-00