Open Framework, Information Management Strategy & Collaborative Governance | Data & Social Methodology - MIKE2.0 Methodology
Wiki Home
Collapse Expand Close

Members
Collapse Expand Close

To join, please contact us.

Improve MIKE 2.0
Collapse Expand Close
Need somewhere to start? How about the most wanted pages; or the pages we know need more work; or even the stub that somebody else has started, but hasn't been able to finish. Or create a ticket for any issues you have found.

Rationale

From MIKE2.0 Methodology

Share/Save/Bookmark
Jump to: navigation, search

The Rationale helps explain some of the key drivers for the Information Development approach and why information-centric approach is critical for today's complex business solutions.

Contents

Changing focus of IT

Information Development in the 21st Century

The IT Capabilities that will be needed in the future will be different and will be driven by the need for information.

In spite of the claims of easy storage and retrieval of information surrounding the introduction of database concepts beginning in the 1980s, implemented systems, for the most part, have not lived up to the promise. Function and data are still tied up in vertical silos. Change is costly and reuse of function is minimal.

Removing cost from the business while providing flexible access to integrated information about customers and products is still the goal. To achieve this, a more formal and yet flexible approach to ‘Information Development’ is needed than what is being done today. This is needed as the definition of an “application” will continue to change as architectures become more federated and more integrated.

The MIKE2.0 Methodology provides a progressive approach to ‘preparing for the future’. It provides a solution that will be delivered over time and be complimented by emerging standards in business and technology. Information Development activities should become part of any organisation’s software delivery approach Mobilisation should begin now; it will take time to prepare and require iterative, low-risk steps during the preparation period.

Data Integration is very complex

Breaking out of the Integration 'Spider Web'

Business Integration Technologies still have some gaps when using them for Data Synchronisation

Business Integration (BPM/EAI) technologies are off-the-shelf integration technologies that are used primarily for event-based integration. They have generally been more successful when used for Process and Application Integration than when used in implementations focused on Data Integration. Failures in the use of EAI technologies is because the problems associated with Data Integration are different than those of Application Integration and most implementations using EAI technologies have taken a Process Integration-focused approach.

Enterprise Information Integration (EII) is a relatively new perspective on the use of EAI technologies. An EII approach uses many of the same technology components underpinning current EAI implementations, but in a manner that is optimised for data synchronisation as opposed to application or process integration. It complements the use off-the-shelf implementation strategies where systems typically house their own data and facilitates a ‘source of truth’ across the enterprise.

The major gaps in most EAI implementations that are addressed by an EII approach are:

  • EAI allows for ‘surface collaboration’ between systems. It is focused only on ‘data swapping’ relevant to required application functions.
  • EAI does not require an abstraction of system-specifics into an enterprise-wide view of information via ‘enterprise business objects’. EII mandates the use of a standardized model for information exchange.
  • EAI typically does take a focus on data quality. An EII approach implements Data Management Services for providing functions such as standardisation, de-duplication, and semantic matching.
  • EAI technologies are also generally inappropriate for handling large, batch-oriented transactions. An EII approach involves the use of complementary integration technologies (such as ETL technologies) that are designed for handling larger transactions.
  • An EII approach ensures the capture of specific business rules that facilitate data synchronisation (e.g. rules for data ownership).

Delivering capabilities for Enterprise Information Integration through a unified integration framework is a key aspect of the SAFE Architecture. It was, in fact, the initiating business driver for creation of the initial solution framework in November 2001.

New challenges introduced by packaged software

There are lots of advantages to using off-the-shelf software, but it does seem to introduce a few new problems

Software vendors have responded to the demand for rapid change with a flood of off-the-shelf applications and integration platforms. Whilst enabling application development and integration to be conducted much more quickly, it has introduced new issues:

  • Off-the-shelf applications largely want to operate on data within their own proprietary data store. This has been the case with custom OLTP applications but is exacerbated by the pre-defined capabilities of off-the-shelf systems that have made data ownership and mediation issues more complicated.
  • Release longevity must now take into account vendor product lifecycles. Even properly working software must sometimes be modified due to support arrangements.
  • A proper approach to Enterprise Architecture is often obstructed by vendor influence.

And of course, many of the ‘old’ problems still exist:

  • Data quality issues still proliferate everywhere
  • Everyone agrees that a strategy for metadata management is important, but its often left out of the project implementation
  • Agreement on common standards is difficult and software components are often not built in a manner that facilitates reusability.

Our objective is not to take a contrarian approach to the current market trends in software implementation. Off-the-shelf applications and integration software provide a number of significant advantages that facilitate a smoother road for project implementations. When implemented in accordance with business requirements these advantages far outweigh any issues.

The point is that in the context of implementing new systems, taking an approach that focuses only on Application Development and Application Integration ensures an incomplete solution. This incompleteness does not necessarily indicate that the solution won’t work – but that problems and inefficiencies will proliferate. In reality it is precisely the areas targeted by Information Development (exposing the shadow process and business rules, determining the data transformation rules between systems, developing standards for information exchange) that become the areas of greatest difficulty when moving from the legacy environment.

In summary, we believe meaningful, cost-effective success (and new Application Development) can only be achieved with the right approach to Information Development.

Cost Reduction

Companies spend too much money on IT to consider a proper architecture design just a “technology problem”

Companies spend an enormous amount of money on Information Technology; with integration usually being largest component of this cost. The changing business environment and constantly evolving technology landscape can render even recent IT investments obsolete.

Although change and rework will always be necessary, a Services Oriented Architecture built around open standards-technology provides the best opportunity for build an Integration Infrastructure than will be reusable into the future. Solutions built as per the SAFE Framework are more flexible, easier to manage, and more accommodating to changes from software vendors.

System replacements programmes are data-dependent

The Stealth Layer for Information Development

The killer of consolidation and replacement programmes are all those ‘little’ reporting systems (that people use to run the business)

Across the information management environment there is a "stealth" layer of data marts and flat files used for reporting. These shared repositories typically support reporting and analysis across the enterprise. Typically they are populated by hundreds of downstream feeds from production environments. Most of these downstream feeds have just evolved over time with little documentation to support maintenance.

These information silos are often times a major barrier in transitioning from legacy systems to new production environments. This is because decommissioning the systems involves more than just the data in the source system - there are now a number of dependent systems that use this core data in different fashions. These dependencies make planning and execution of decommissioning very complex and often risky from a business continuity perspective.

The Information Development approach ensures that the stealth layer is addressed during the transition and is not replicated in the new environment.

We have moved out of the batch world

BusinessTime Model for Information Development

Delays in information are becoming less and less acceptable by the business

Most business requirements do not mandate transactions be done in batch; limitations in legacy technology have driven these design decisions. Improvements in integration technology now allow the real-time integration of information to be done more easily, at the same time the shortening cycle-times for business decisions and the increase in integration between distributed systems pushes the need for more real-time, event-based data synchronisation.

We believe you make use of a BusinessTime model when defining requirements. This simple model classifies when synchronization of business events are required by different systems to meet their information needs.

Wiki Contributors
Collapse Expand Close