When data modeling for an operational data system, what consideration is given when developing the logical data model for supporting data integration with existing data systems?  When I learned data modeling, many years ago, integrating the developing data model with other existing data models was not an issue to consider.  Has this changed or is the integration of data models still not a consideration in the methodology?

Is data naturally integrated or does it naturally exist as islands of disparate data?  Our investigated in search of the natural state of data began with our universe of data.  Our universe of data is composed from all metadata and all data from all data sources. Data models are perhaps our best  depiction of our universe of data. Within each well-developed data model, every data entity is connected by entity relationships to other data entities.  No data entity or cluster of data entities is isolated or partitioned from any other cluster of data entities.  There are no islands of disparate data represented within a single data model and there is no sign of data isolation.   

While we have certainly developed islands of disparate data within our data architectures, data integration methods may be applied to integrate this isolated data. Since any island of disparate data may be resolved by current data integration methods, one could therefore conclude that our universe of data should be integrated.  That is to say that the natural state of data is to be integrated.

So why is data isolation so prevalent in our data architectures?  We have discovered
that these islands of disparate data are an unintended artifact of our currently flawed data modeling methodology that ignores the interactions between data  models.  With current data modeling methodologies, data models are developed independently and composed from heterogeneous metadata.  The lack of metadata commonality and the lack of links between data models result in metadata and data isolation.  Since each database is normally developed from a single data model, the resulting database is just as isolated as the original data models.

Our new Data Reintegration Methodology corrects the database design flaws and eliminates the data isolation that was instantiated into our databases. Our data models now support relationships between data models. This is another first in the industry! These inter-data model entity relationships are named commonality relationships.  Commonality relationships, when properly instantiated, supports referential integrity between databases.  Within the Data Reintegration data architecture, each data model becomes a part of a distributed network on integrated data models.  When databases are instantiated from these data models, each database becomes a part of a single distributed network of integrated databases. 

You are invited to leave comments on this blog.  For more detailed information, 
contact us and request the Data Reintegration Methodology whitepaper.

Data Reintegration is a trademark of Strategic  Insights Inc. 
The Data Reintegration Methodology is patented by U.S. patent no. 7,979,475 and other pending patents. 
©Copyright Strategic Insights, Inc. 2012.  All rights reserved.

Our universe of data is composed from all metadata and all data from all data sources.  Have you ever envisioned the major properties of this universe of data?  Well, don’t feel too bad if you haven't as you certainly are not alone!  However, having a universe of integrated data is a very big deal and the boundary of master data plays an  extremely import part.

Data models are perhaps our best depiction of our universe of data.  Every data model has a boundary which is composed of master data entities.  However, in examining multiple data models, each data model appears to have a different boundary of data entities. Our research in this area has established that our universe of data is finite and therefore does indeed have an actual boundary of master data.

All current state of the art data models are not based upon a common set of boundary data entities.   Metadata commonality does not exist between these isolated data  models. One could envision all current data models afloat and lost in our universe of data.  Current data models do not accurately reflect the properties of our universe of data.  These data models are not properly anchored, and certainly do not support integrated data.   

The Data Reintegration data architecture accurately reflects the properties of our universe of data.  The Data Reintegration Methodology adds actual boundary data entities to each data model which anchors that data model within our universe of data. More importantly, each data model is now anchored relative to each other data model.  These actual boundary data entities in each Data Reintegration data model are the foundation of metadata commonality linking data models together. These links are referred to as commonality relationships in the Data Reintegration Methodology.  An  instantiated Data Reintegration database has referential integrity with all other Data Reintegration databases that share metadata commonality.  Each Data Reintegration data model becomes part of the single distributed network of integrated data models.  Each Data Reintegration database becomes part of the single distributed network of integrated databases.

You are invited to leave comments on this blog.  For more detailed information,  contact us and request the Data Reintegration Methodology whitepaper.

Data Reintegration is a trademark of Strategic Insights Inc. 
The Data Reintegration Methodology is patented by U.S. patent no. 7,979,475 and other pending patents. 
©Copyright Strategic Insights, Inc. 2012.  All rights  reserved.

The reason databases currently are not data integration friendly is simply because they are not design to be data integration friendly!  Currently, data integration is not a database design consideration.  Most data models are independently designed and result in the definition of  heterogeneous disparate data models.  Each independently designed heterogeneous data model, when instantiated as a database, forms an disparate island of disparate data or if you wish an independent information silo.  For example, ten heterogeneous data models, when instantiated, result in ten isolated islands of disparate data.  These islands of disparate data and information silos are dominant in our current data architectures.  However, the root cause of this data isolation has not been previously defined.
 
Without understanding the root cause of this data isolation, several data integration methods have been developed over the past 25 years.  Now that we understand the root cause of this data isolation, we know that none of the prior art data integration methodologies correct the actual problem.  Since it is our database design methods of defining disparate heterogeneous data models that cause data isolation in our data architectures, the correction of the root cause needs to be a correction of the data modeling methods.

The Data Reintegration Methodology is a long overdue modification to the current data modeling methodology.  Amazingly, eliminating the data isolation caused by our heterogeneous data models results in the integration of these previously isolated data models.  That is, all heterogeneous data models that are created or enhanced using our patented Data Reintegration methodology become a part of a single network of integration data models.  Likewise, the databases instantiated from the integrated data models are also integrated provided that these databases are properly populated with data as defined within the Data Reintegration Methodology.  Under these conditions, every Data Reintegration database becomes a part of a single network of integrated databases.  Any database that is enhanced using the Data Reintegration methodology is now integrated with any other so enhanced database.  With the Data Reintegration Methodology, we are simply removing the data isolation artifact of our prior art data modeling and permanently re-integrating the data that never should have been isolated!

You are invited to leave comments on this blog.  For more detailed information, contact us and request the Data Reintegration Methodology whitepaper.

Data Reintegration is a trademark of Strategic Insights Inc. 
The Data Reintegration Methodology is patented by U.S. patent no. 7,979,475 and other pending patents. 
©Copyright Strategic Insights, Inc. 2012.  All rights reserved.