Problem solve Get help with specific problems with your technologies, process and projects.

IBM system z9 and information integration: What's the connection?

IBM system z9 and information integration: What's the connection?

Two significant events have happened recently:

  1. IBM has rationalized its data-related middleware and put it under the heading of "information integration," as part of an "information on demand" strategy.
  2. The IBM System z9 folks have begun a strong strategy of positioning the System z9 as the premier platform for data serving.

How do the two relate? Or, to put it another way, how can IBM customers leverage both information integration and the System z9's new data-serving capabilities?

Information integration

IBM's information integration solution set is less about the name and more about what customers can do with IBM's new data-related software architecture.

That architecture consists of individually useful capabilities, as shown below.

Table 1: Taxonomy of key information integration tools

Tool aTasks Uses
Replication, EAI, and ETL tools

Example: Those in WebSphere DataStorage (formerly Ascential) & WebSphere Information Integrator

  • copies one data piece at a time, one way, either continuously as data changes or in packets (typcially offline)
  • Can transform data to the different format of a data warehouse and "cleans" the data for higher quality
  • Typically many sources, one target
  • Update a data warehouse or data mart with new or changed data from business-critical OLTP relational databases, often hourly, daily, or weekly
  • Reflect local-office data to a central site (e.g., backup or combining the data for reporting)
  • Cleanse data to assure its quality and consistency
  • Copy data between packaged applications, e.g., to integrate business processes (Enterprise Application Integration)
Database as a data warehouse or data marts

Example: DB2 for z/OS

  • Receives copies of large amounts of data, often offline (data marts may copy to data warehouse)
  • Typically many sources, one target
  • Allows querying of combined data but not update
  • Often very large (terabytes)
  • Support querying, data mining, and business intelligence
  • Provide metadata repository of key business information
Database as an operational data store (ODS)

Example: DB2 for z/OS

  • Copies (one way), typically changes in data (immediately)  often from the store/cache to multiple targets, always online
  • Usually many sources, one target
  • Often uses replication or ETL tool
  • Allows querying and update
  • Often large, but not as large as data warehouse
  • Combine customer and supplier information to detect relationships and use for competitive advantage
  • Combine business-critical application data for business process integration
  • Improve time-to-value and speed-to-react by seeing more up-to-date data than data warehouse can provide
  • Improve visibility of current (less than a day old) information
Enterprise Information Integration (EII)

Example: WebSphere Information Integrator

  • Preserves the data sources as they are
  • Often uses real-time ETL tool or "adapter"
  • Allows querying and (in some cases) update
  • Integrates data and content and views as a single source
  • Supports queries/reads/updates across a broad range of data sources (including file and rich-media data) without creating copies to synchronize
  • Creates a more comprehensive metadata repository than any other strategic information management tool
  • Allows development that includes cross-data-source transactions
  • Potentially, supports cross-data-source administration
Source: Infostructure Associates, January 2006

Combined, these information integration tools, on any platform, can play a key role in business-strategy success, as shown in Table 2. The "VOS" in this Table is a virtual operational store, combining ODS and EII, which acts as a "virtual" enterprise-wide database.

Table 2: Ways to use information integration tools in key IT and business strategies

Aim Ways to use a VOS
Integrate business processes
  • Replicate data from the data sources involved in the business processes to the ODS, allowing all related transactions in the business processes to be carried out at once within the ODS in a consistent fashion  and then the resulting data is resynchronized with the original data sources.
  • Persist data views aggregated across applications within the ODS for improved query and update performance across the applications
  • Program using the EII interface rather than the component applications, in order to carry out querying across business processes. The resulting code should use the EII tool's ability to combine data from different sources and allow analysis of data relationships across data sources, not just view data from different source side-by-side on the same screen.
Implement Web services interfaces to existing information
  • Write Web services consumer code for each key enterprise application to invoke the VOS rather than each data source
Improve RTE capabilities by speeding time to react to current customer or supply chain data
  • Replicate high-frequency-of-update customer or supply-chain data in the ODS
  • Use the VOS to offload processing from back-end enterprise data stores and systems of record, speeding notification of new data and changes caused by reacting to new data. Leverage the VOS's extensibility to implement changes to a broader array of key data.
Reduce time to develop new applications leveraging proprietary information, such as portals
  • Use an EII tool to front-end the data warehouse plus other proprietary information, allowing one set of code for all information access
  • Use data persistence in the ODS to reduce the need for mapping and reconstruction programs for unstructured content
Audit all information in the enterprise for government requirements or Sarbanes-Oxley
  • Create a metadata repository in the VOS by importing or generating metadata from data warehouses/marts, ETL tools, and file systems and rich-media sources. Write audit queries to the resulting cross-data-source EII interface, or create a reporting application with modular report templates and reusable components.
Cut IT costs
  • Cut development costs via the VOS's ability to allow developers to "write once" to supply data access to many applications
Attain better scalability of existing multi-tier applications
  • Implement a VOS to aggregate information (both structured and unstructured) from back-end systems of record and store it in the operational store. Synchronize the aggregated data views with back-end systems via the VOS. Thus, the user can offload query processing from back-end systems and satisfy the data requirements of multi-tier front-end applications with higher performance.
Facilitate mergers and acquisitions
  • Using EII, create integrated reports about customers, suppliers, and the overall enterprise's financial position, e.g., over multiple data warehouses/marts.
Infostructure Associates, January 2006

System z9 data serving

System z9's new data-serving capabilities, as announced on 1/24/06, have three components:

  1. Improvements in DB2 on the System z9.
  2. Improvements in other data-handling capabilities on the System z9 (zIIP).
  3. Improvements in other System z9 capabilities that affect data-handling, e.g. data security.

DB2 for z/OS improvements, in V8 and the forthcoming "vNext", include:

  • Support for high-performance "native" SQL stored procedures.
  • Extension of database security via "roles" and auditing.
  • Performance boosts, e.g., an Optimization Service Center.
  • Simplification, e.g., via "partitioning by growth" and default databases/tablespaces.
  • Service Oriented Architecture (SOA) support, including integrated XML data and WebSphere integration.

Of these, perhaps the most important is the support for a "superset" of standard SQL, so that most applications can access DB2 for z/OS (and IMS) or other databases with the same code.

zIIP is a specialty engine for the System z9 that allows outside applications to access mainframe data stores with high performance and the rest of the System z9's virtues. As IBM notes, zIIP allows such non-mainframe applications as ERP, CRM, BI, and data warehousing to access and take advantage of business-critical mainframe data much more easily. This, in turn, gives users a good reason to place more of their data centrally in a common repository, on the mainframe data server.

Other key data-serving features in system z9-109 include:

  • Greater performance from the hardware (e.g., 80% more bandwidth and 40% more FICON channels, leading to up to 125% improvement in I/O and 56% improvement in response times).
  • Robustness features, include redundant I/O interconnects and Parallel Sysplex.
  • Business compliance and security features, including higher-performance advanced encryption with protection for data shared outside the organization.

Marrying the two

The need to unite mainframe data serving and information integration stems from two long-time facts:

  1. Much of a typical large enterprise's structured (numeric) mission-critical and business-critical data continues to reside on the mainframe.
  2. Much of a typical enterprise's semi-structured (e.g., text) and unstructured (e.g., graphics) data that relates to business-critical data — for example, a hospital's charts and doctors' notes relating to a patient — resides outside the mainframe.

Until now, no one could do anything about this, for two reasons:

  1. The information integration tools that allow a broad view of data across both mainframe and non-mainframe data stores had not yet been created or integrated effectively.
  2. The mainframe did not support standard transactional access such as standard SQL.

Thus, the two events cited above are key to marrying information integration and mainframe data serving:

  1. IBM's new information integration product suite allows users to issue transactions easily across mainframe and non-mainframe data stores.
  2. IBM's new and announced System z9 capabilities not only allow applications to see the mainframe as similar to Unix/Linux and Windows servers (via encapsulation of mainframe applications as services and support for standard SQL) but also make the mainframe a strong platform for carrying out all types of transactions (e.g., XML support, performance enhancements, and security and robustness features).

In other words, using IBM's information integration tools and the latest System z9 technology, a user can much more easily "redirect" a business-critical application such as ERP, CRM, or BI to use mainframe as well as non-mainframe data, and store structured, semi-structured, and non-structured data equally on the mainframe or off it.

Fruitful areas to apply system z9 plus information integration

The marriage of information integration and System z9 also offers users the opportunity to employ a mainframe in new ways.

For example, users with mainframe data warehouses can now add semi-structured and unstructured data to these. Thus, the "extended data warehouse" that is often discussed these days will support more current data gathered from a greater array of data sources such as outside the organization on the Web, and BI applications that before could not access the mainframe's data or were forced to copy it frequently to another platform can gain richer insights with higher performance.

As another example, users in such efforts as "master data management" are now seeking to create metadata ("data about data") that establishes the relationships between all copies of related customer, supply-chain vendor, or distributor data. Much of this key data is on the mainframe; therefore, for performance's or simplicity's sake, it makes sense to place the repository of this metadata on the mainframe. The mainframe can therefore potentially be used as the focal point for an effort to rationalize the data architecture of the enterprise, with dramatic effects on IT costs, the ability to gain strategic insights, and react to events rapidly — "information on demand."


Of course, not all difficulties have been swept away. Customers should expect to have to carry out due diligence, formulate a process for incorporating key applications in the mainframe-information-integration marriage, and deal with the gritty details of implementation.

Nevertheless, I conclude that a mainframe-information-integration marriage is now, for the first time, real, and its potential achievable. In turn, that means that the mainframe is of greater strategic importance to the enterprise beyond its own boundaries, as an integral and by no means negligible part of overall IT information efforts. New ways to leverage mainframe data serving using information integration are nice for the customer; but leveraging the mainframe as part of an overall information strategy is even nicer.

About the author:Wayne Kernochan is president of Infostructure Associates, LLC, a Lexington, Mass.-based analyst firm. Send Kernochan an e-mail here

Dig Deeper on IBM system z and mainframe systems

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.