Home > Uncategorized > Interesting view on non-relational database management systems alternatives

Interesting view on non-relational database management systems alternatives

Is the Relational Database Doomed? by Tony Bain

As a matter of fact, up untill probably couple of weeks ago I N-E-V-E-R heard anything about non-relational database management systems. Not that it’s automatically supposed to mean they never existed, – obviously not, but  definitely implies there are far, far not that popular as RDBMS. 

This notion of non-relational DBMS alternatives a bit kicked me to the head and made me decide to write this short critical post.

At first, I must admit, the author does make his point, but to a very small degree. There are quite a number of serious flows in his argument, making it totaly weak.

It starts with the most minor one right at the very beginning    😉 But that’s nothing really, of course. “Fixing” it just for fun:

–Civic is Honda
UPDATE MakeModel
SET MakeKey = 1
WHERE MakeKey = (SELECT MakeKey FROM Make WHERE Make = ‘Nissan’)
AND Model = ‘Bluebird’;

Then the author mentions that 

“new” type of database has been around for a long time and has been used for specialized applications for which the generic relational database was ill-suited

without giving a single example… Not a cool thing to do, especially attempting to question such a serious matter as  prevalence of relational DBMS over any other database systems and, what’s more to it, putting the entire future of RDBMS to a subject. One would expect some really kind of a bullet-proof, very obvious, no-flaws argument in this.

Then comes something weird. Well, I’ve modified it a bit – the way it supposed to be.

The data model is based on business requirements. Application functionality is supposed to reflect the very same business requirements and follow the data model, if this is relevant to mention at all. This entire thing doesn’t make any sense. The next one, about data model being normalized is true only for some systems, called “Operational” or OLTP in the older notation but not Data Warehouse ones. Data Warehouse systems has data de-normalization as its very base principle. And further on in his article, the author keeps building his argument like Data Warehouses never existed in relational database world – WTF?!?!

 The rest of his argument goes nowhere really. Such features as parallel execution, partitioning and data compression, which make up great performance improvements in relational databases, have never been mentioned in the article. At the bottom line, his point is this: “if you want to go with WebServices, choose key/value non-relational DB”. Well, cool, dude. Do it if you dare, – it’s “your” money at the end of a day. Well, not really yours, but the corporate ones. It’s all the question of a what you’re willing to risk.

All in all, the time and the market will show whether those “key/value” barely known around database systems have their potential or not. Meanwhile, relational databases will be there as the one and the only main player for a lifetime of computers operating on the basis of zeroes and ones.

  1. No comments yet.
  1. No trackbacks yet.

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: