Business intelligence tools have taken the lead on solving the thorny problems related to cross-platform data integration. Our expert explores the options available to IT organizations.
Some might argue that IT pundits are expounding on the obvious when they say, "Everything is converging." But business intelligence, at least, is doing so in a unique way.
Our recent article on BI, business activity monitoring (BAM) and business process management (BPM) ("Where Process And Intelligence Converge," January 2005) described the movement of BI from strategic heights into operational details, such as real-time business activity monitoring and business process management. Naturally, these systems get very close to performing basic data-processing tasks. For example, enterprise resource planning (ERP), customer relationship management (CRM), supply chain management and other enterprise systems lend vital data feeds to BAM and BPM systems. BI and BI-related vendors clearly are now stepping into and appropriating data initially captured for basic operational purposes.
Many of those data sources, residing within tools from vendors like SAP, Oracle, Siebel and others, are now returning the compliment. They are saying, essentially, "We can provide the data stewardship services that BI and BPM vendors provide, plus the BI analytical processing."
As the promise of information at users' fingertips becomes more and more of a deliverable reality, the battle is on to decide which underlying software packages will deliver such capabilities at every level in the organization. As in so many areas of computing, the Battle of Convergence and Consolidation is shaping up in business intelligence. In this article, we look at BI, enterprise application integration (EAI) and BPM to see what skills, tools and resources each brings to the table.
BI Tools' Role In Data Integration
Business intelligence applications bring a number of important capabilities to the data integration and interoperability table:
Visual design and layout tools that handle objects, images and animations
Output to a broad range of formats: CSV, Excel, HTML, Flash, PDF, SVG, XML and others
Output that is cross-platform and customizable for styling, look and feel
Data staging, hubs and warehousing functionality, including snapshots and versioning
The ability to respond to real-time events and provide data updates, replication, and dynamic maps as specified by business rules
The ability to deliver the right info, to the right place, at the right time and in the right formats
In recent years, BI has moved to a much more cross-platform base. Client applications have broadened from being primarily Windows-based, while output and processing have also become more open and cross-device. The underlying data integration infrastructure formed by data warehouses and ETL systems has been made better through event-monitoring.
EII And EAI Tools
It's no surprise that many enterprise application integration and enterprise information integration (EII) vendors -- Avaki, Attunity, Composite and Kalido, for example -- have strong ties both to each other and to the BI and BPM communities. EII is an outgrowth of the old mainframe canard that every organization needs a master data repository. Such a central repository helps to enforce data-coding standards and data interchanges. Current EII systems also add one crucial capability -- the ability to create federated views of distributed data across different databases, files, and other structured data sources scattered throughout an organization. That's no small task. Even better, some EII systems, when presented with a query, help determine an optimum query strategy given the need to transfer varying quantities of data over connections of differing speed, reliability and concurrency.
EAI, on the other hand, builds connections between various application stacks, knitting an ever wider cloth of data and process interoperability. EAI also uses the broadest set of technologies to deliver its integration services. This is where the hype of simple object access protocol (SOAP) and Web Services gives way to more tested approaches like common object request broker architecture (CORBA), message-oriented middleware (MOM), J2EE, XML-RPC and proprietary applications.
EAI's benefits include:
Features broadest set of cross-platform technologies
Ties in with a wide range of app gateways: J2EE, .NET, Mail, Directory, etc.
Has robust set of structured data mappers and movers
Works both synchronously and asynchronously with appropriate monitors
Many are metadata- and business rules-aware;
So EII and EAI are on the front lines of application and data integration. But even with structured data, there is no single best way to tackle all data integration tasks. Web Services offer the hope of making data, methods and objects available from anywhere, but face daunting authorization and security risks.