I have to pick you up on scaling.  U2 scales really well there are some 
staggering performance test that have been done.  U2 does things differently, 
but it does scale.   An area that NOSQL databases are not demonstrating in what 
I have looked at, is the transactional processing and security that a 
commercial database requires.  It is not to say that Rocket does not have some 
work to do in some areas, but I don't get that something new is the be all and 
end all and that something that is mature is outdated.  I like to try and get 
the best out of both.

David Jordan

-----Original Message-----
From: u2-users-boun...@listserver.u2ug.org 
[mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of DavidJMurray 
Sent: Saturday, 17 December 2011 1:38 PM
To: u2-users@listserver.u2ug.org
Subject: Re: [U2] Graph DataBase

I also, to some extent, agree with Dawn.

The mv, NF**2, post relational, PICK, or whatever the correct model name is, it 
is not a relational database. There are very strict definitions to a relational 
database, namely, but not complete, normalisation, SQL, joins etc.

Yes, the mv model can emulate a relational database. But, it you throw away the 
dict part of a file, the data is actually in a form of a document.

And, if you go way back to the definition of a database model created for 
analysis - Entity-Relationship by Chen in the early 1980's, the mv model is 
very close to the ER model. So, if you add back the dictionaries, the database 
model could be a DDERDBMS - Dictionary Driven Entity-Relationship Data Base 
Management System.

Anyway, whatever mongoDB classify themselves as, so should the PICK model.

But then, U2 is not that well known, as it is not marketed to end users per se. 
It is an embedded database for VAR's. The focus is on the application, not the 
underlying database. This could be seen as a problem in some markets. So, it 
could be seen to be lost energy promoting the database to end users or 
developers within the nosql market as U2 is not open source, hence costs money, 
does not scale that well (sharding and/or federation) and is not really for 
high volume loads within a web service delivery environment.

It's a good solid product for well established actively-promoted vertical 
market applications which need to extend to the surrounding I.T. ecosystem.

Dawn Wolthuis wrote:
> I do disagree with this. If neo4j meets the not-determined criteria 
> (except by marketing departments) then MV products do too. MV  vendors 
> might have suggested they are relational at varying points in their 
> history (again marketing depts) but they do not meet Relational or at 
> least SQL-only DBMS criteria. We are very much in the not-SQL-only 
> camp whether we proclaim it or not. No product in that mix does 
> everything. I put Tom del's nodal logo on a blog entry in 2006. It is 
> someone in the MV space that has the nodal domains. We are NoSQL, no 
> doubt, but we can decide to stay in our MV sandbox rather than joining 
> the game if we so choose. --dawn


Learn and Do
Excel and Share

http://mvdbs.com http://mvdbs.com
View this message in context: 
Sent from the U2 - Users mailing list archive at Nabble.com.

U2-Users mailing list
U2-Users mailing list

Reply via email to