Current Blog Entries by Larry Fry, CCP, MBA

Another WordPress weblog for my Business, Energy, Investing, IT, and Travel blog entries!

Status of Object-Oriented Database Management Systems?

with one comment

Being a senior business systems (IT) consultant who has always utilized hierarchical (IMS) and relational database management systems (RDBMS), I’m wondering whatever happened to the concept of object-oriented database management systems (ODBMS), which at one point (i.e., 10 years ago) was considered to be the database structure design of the future by some IT academic types for object-oriented software applications?  Do the newer data storage mechanisms and/or models currently out on the market represent various hybrid versions of ODBMS?  Or has this concept completely gone by the wayside before it ever had a chance to take off due to its being considered “too disruptive” of a technology by IT management?  My impression has always been that an ODBMS persistence structure would need to be in place in order for a true object-oriented software application to be able to make use of (and persist) intensive object nesting, cross-referencing, data type extensions, sub-classing, and inheritance support functions. But perhaps the more powerful RDBMS engines and the extended versions of the same have now enabled object-oriented software applications to circumvent some of the traditional performance and storage issues that traditionally have occurred with RDBMS’s. This would also include circumventing some of the object-to-relational and relational-to-object translation mapping issues that traditionally occur when persisting/recovering software objects using an RDBMS. This translation mapping issue occurs here because the developer has to break the objects apart and then store the different object parts in the various database tables (or files) making up the RDBMS, thus making the translation mapping process tedious at best due to the resulting “semantic gaps” that occur. In any case, my assumption is that providing for the full functionality of an ODBMS, which would also enable developers to persist full class information (including object methods), is the biggest hurdle for the ODBMS genre to try to pass, thus rendering the ODBMS niche as being “too disruptive” of a technology for most IT management types.

Interested LinkedIn members are invited to join the ”Disruptive Technologies” professional group (URL below):

http://www.linkedin.com/groups?about=&gid=1027037&trk=anet_ug_grppro

Advertisements

One Response

Subscribe to comments with RSS.

  1. ODBMSs are alive and well. They’re still serving niche markets because RDBMSs are perfectly adequate for most standard tasks. Objectivity and Versant reported record financial results last year. Interestingly, the open source contender, db4objects, got taken over. However, so did MySQL.

    Storing methods in the database never really got off the ground, mainly because of heterogeneous language environments. Users didn’t want to get locked into proprietary environments.

    Efficient storage and traversal of relationships has proven to be a strong point. In a recent benchmark the RDBMS needed 300 TB to store 100 TB of raw data and the ODBMS needed 120 TB. Take a look at the link hunting demonstration featured on the Objectivity site. It takes 30 to 40 seconds to perform a task that was taking a leading RDBMS forty to sixty minutes.

    The ODBMS companies will be at the International Conference On Object Oriented Databases 2009 in Zurich, Switzerland on July 1-3.

    Leon Guzenda

    June 14, 2009 at 9:44 am


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: