Managing Complexity with EA

by | May 16, 2008

Over time, technology provides more and more functions and hopefully value to your enterprise. The challenge is how to manage the complexity that comes with technology. I started my IT career as an IBM 360 mainframe computer operator managing VM/CMS and DOS/VSE CICS systems. IT architectures were relatively simple. One large computer, a few large boxes for hard disk drives, several tape drives and an a large line printer. For the most part, our clients interacted with printouts and some lucky ones got access to 3270 green screen terminals.

Think about today … while our technology provides a very functionally rich environment for delivering value to our organizations, the technological complexity has gone through the roof.

Delivering our ERP in 1992 took one IBM RS/6000 RISC box with Oracle 6 RDBMS and Oracle forms installed on all client PCs. Now we need a P Series Server running AIX (Unix) for the Oracle database, VMWare Servers running SUSE (Linux) for the Oracle Application Server(s), Load Balancers, DNS servers, DHCP Servers, Novell network servers and various workstations (PCs running Windows XP and IE 6 or 7 with client side Java)!!!

So how do we as Enterprise Architects manage complexity? We created an Application Portfolio with attributes about the components that the applicaiton requires like:

  • identity store
  • web enabled
  • database server o/s
  • application server o/s
  • web server o/s
  • web server
  • database
  • application development environment
  • etc ..

We have over 200 applications with at least 7 distinct solution delivery platforms:

  • Oracle – J2EE
  • LAMP
  • Microsoft – .Net
  • Lotus Domino
  • Sun
  • Apple
  • Novell

Now we are working on the second year of our Technology Plan (rolling 3 year window). As part of this we will be looking to favour “depth in our technology choices over breadth”. We will focus on Oracle, LAMP and Microsoft. Investing in these solution platforms will allow us to focus our people on a managed set of technologies and be able to more rapidly respond to our clients needs.

Anyone else thinking “Deep instead of Wide”??

5 thoughts on “Managing Complexity with EA

  1. T-Enterprise

    Excellent article – we use something similair to your app portfolio however not as structured nor as many delivery platforms – will get there!

  2. FEAC

    Calling what your are describing Enterprise Architecture is totally different from what EA really is. EA is a discipline that allows us to get our arms around complexity in the 21st century. Yes you can apply that discipline to the technical problems you talk about but it is much larger than that. It is an all encompassing holistic discipline that gets you from strategic planning all the way to instantiations and better more real time 21st century decision making. Using it in the context you describe misleads people into thinking it is all about technology. Just the opposite is true, its about the business first and then how technology can help make the business more agile and more profitable.

  3. LeodeSousa Post author

    Felix, Thanks for the comment … a bit harsh but I am good with the feedback so that I can improve on the work I do.

    First, I disagree with your comment that what I posted in this blog entry “totally different from what EA really is”. I suggest that if you want to know how we have built our EA practice, you need to go deeper than reviewing one of my blog posts about complexity and condemning me with “misleads people into thinking it is all about technology”.

    Our EA practice is guided by the following:
    – “Do the Right Things” – this is Governance with EA
    – “Do Things Right” – this is EA with other methodologies like PM and ITIL

    The post above was about “Do Things Right” …

    Regards, Leo

  4. John Wu

    Hi !
    LiteEA believe EA is both wide and deep. EA is a discipline see the forest but also eye on the tree to look for reuse and consolidation. We look at big picture to see the forest and eye on the depth to find the commonality and enable agility. My recent article on illustrates the EA evolution from the big bang approach, the business centric approach and the incremental approach. On the illustration, you can see the wide branch high level planning and detail branch of the reuse and consolidation. EA has evolved in the two direction of business centric EA and the incremental EA. It is not for us to say who is right or wrong, only time can tell. However, the business centric approach has been a continuous struggle from the fortune tellers in the early days to the industry engineering, enterprise engineering. Hopefully, EA do not become the same struggle.

  5. Felix

    You are right Leo I was only addressing what is written above. So my view was limited to that context only


Leave a Reply

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

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