Special thanks to Loraine Lawson for pointing out this excellent article in DMReview titled “BI, Process, and Integration Trends.”

The article is clearly written with the audience of Business Intelligence (BI) specialists in mind.  However, it has some excellent points to make that Enterprise SOA folks will see value in. Highlights for a SOA audience:

“Key elements, which BI [Information Management] share with [Enteprise Integration Management], are data integration and process. Data integration is the technology enabler, while process addresses the authority and management of the information asset.

“The success of enterprise data integration hinges upon leveraging a mature information architecture and integration practices prevalent in BI, while evolving to account for recent architectural concepts, such as service-oriented architecture (SOA), to ensure the hunger for timely and meaningful information is satisfied.

The authors go on to list a series of challenges that are shared by BOTH integration and BI spaces.  I agree with this list wholeheartedly.  I have only listed the items.  The full article goes into a bit more detail on each.

  • Address the lack of a global data model.
  • Debate of using BI as an operational data store (ODS).
  • Define how and where to implement master/metadata; data cleansing, business/transformation rules.
  • Lack of data quality management within legacy and in-house applications.
  • Define an enterprise data integration architecture.
  • Acknowledge technology disruption from Web 2.0/SOA.
  • When selecting the right long-term partnership with vendors, one must look for the right mix of capabilities.

I’ll end this summary with another excerpt that I think rings loud and clear to a SOA audience:

“BI and SOA architecture disciplines support similar needs: IM, metadata, data integration and data quality. There is an education process that must occur between the technology camps (and integration vendors) to ensure each understand and exploit the synergies of both architectures. At this time, tools and opinions abound, and the right choices must be rationalized against business values.

Advice: SOA folks: if you aren’t already working with your BI team, pick up the phone.  Their mature processes and practices are able to address many of your issues, and the natural synergy between BI and SOA can make them a strong ally in the fight for a better, faster, cheaper, and more intelligent enterprise.

By Nick Malik

Former CIO and present Strategic Architect, Nick Malik is a Seattle based business and technology advisor with over 30 years of professional experience in management, systems, and technology. He is the co-author of the influential paper "Perspectives on Enterprise Architecture" with Dr. Brian Cameron that effectively defined modern Enterprise Architecture practices, and he is frequent speaker at public gatherings on Enterprise Architecture and related topics. He coauthored a book on Visual Storytelling with Martin Sykes and Mark West titled "Stories That Move Mountains".

6 thoughts on “SOA and BI support the same goals”
  1. @Arnon,

    Excellent Article.  I agree with it completely.  If you look at the article that I linked from DMReview, you will see that the author intended the same notion as you did: SOA+EDA –> BI

    This sentence from your InfoQ article is particularly quotable:

    "EDA on SOA solves the BI problem; as soon as you have event streams on the network, the BI components can grab that data, scrub it as much as they like, and push it to their data marts and data warehouses."

    We are on the same page.

    The reason that I say that SOA and BI have natural synergies has to do with another statement that you made.  

    "ETL with SOA might even be easier, considering that SOA promises that the enterprise data would be woven into a cohesive fabric and not some point-to-point integration spaghetti."

    An Enterprise SOA can ONLY work if the messages being sent, at the enterprise level, derive from a common data model.  The same common data model is required by BI to produce the data staging area that is used to generate ROLAP cubes.  

    Someone, working on behalf of the enterprise, has to create that operational data model.  Both BI and SOA need it.  Team up.  

    Yes, there are natural synergies… in the core understanding of the enterprise needed for both operational integration and business intelligence.

Leave a Reply

Your email address will not be published. Required fields are marked *

five × two =

This site uses Akismet to reduce spam. Learn how your comment data is processed.