Comments on: Robert Greene on “New and Old Data stores” Trends and Information on AI, Big Data, New Data Management Technologies, Data Science and Innovation. Tue, 18 Jun 2019 08:07:11 +0000 hourly 1 By: Robert Mon, 13 Dec 2010 21:25:38 +0000 I am not aware of any standards for the distributed file system, key/value, document, graph database technologies, but for the “ODB”, there is ODMG which a few of the vendors have implemented.

Then there is JDO for the Java space, it is used via Apache/DataNucleus for RDB ( most noteably for Google App Engine ) and it does afford people the opportunity to switch for performance and/or operational reasons from RDB to ODB.

Though granted, there are certainly more JPA users and it would be good to see that API on more ODB’s..albiet with lots of vendor extensions to make it look more like JDO and bring ODB power features back into play.

By: Pieter Tue, 07 Dec 2010 09:20:46 +0000 Just wondering if standards play a part in this field of new types of databases? Maybe it shouldn’t because the db you select is selected for a specific type of problem or application space?

Still wish odbms vendors would invest some time in JPA.

“If it is model driven, I lean towards ODB or RDB.”

Then if it is model driven and I start out with RDB I could easily switch to ODB if I was using a standard interface like JPA.

By: OP Choudhary Sun, 05 Dec 2010 04:13:17 +0000 Robert has sorted through the various DB technologies popping up in recent past. His explanations are simple to understand. Even a business user can grasp well the complicated maze of database landscape. For a long time, there were very few DB choices and in the recent past, so many new options have popped up. The application complexities, data volume are growing exponentially, and technology environment is progressing very fast with the internet speed and perimeter devices like handhelds and PDA’s. I believe that has necessitated the growth of DB vendors other than Oracle!

If you talk to today’s DBA’s and programmers, you will find very few who will know any other DB technology other than Oracle, DB2, MySQL & SQL Server for sure. Anybody has heard VSAM, IMS (hierarchical DB), IDBMS(network DB) and so on? For a long time DB technology progress stopped at the RDBMS platform with few exceptions. ODB was the only new technology which made waves in 1990’s and there were lot of excitement for a very short time. The applications complexity was well handled by existing technologies and the customers did not had any other compelling reasons to adopt this other than new technology and we see a limited adoption of ODB. On the other hand Object Oriented programming was adopted very well. This is where I see a big mismatch of the technology evaluation of 1990’s. We have been dealing with tables, rows, and joins at the database level and objects models at the programming level. Somebody has to bridge the gap and we see the products doing the Object-Relational mapping and introduced another performance lowering activity.

With the modern complex business problems, which stems from appetite for large data needs and complexities of data models, you will see a space for many DB technologies and healthy market potential for ODB’s. Roberto Zicari has done a great job in capturing the Robert’s thoughts and addressing the confusion going on for a long time. Expect more on this topic in the future.