Re: B*-trees

Hi Carl,

thanks for your reply and thanks for the email address and the links. I had a look at your web page and the "query by example"-interface. It's interesting, but i guess if you have a non-objectoriented database it's better to implement some kind of SQL-like script language or some other query language... for my purpose a very simple query language is sufficient.

> Depending on the structure of your data, reading from one file could also be
> faster, if one read operation gets more than one index table.

Sure, the head of the harddrive doesn't have to move that far... now it's clear to me.

> > 3. Is there a kind of "Tiny-SQL"-standard which offers a very reduced
> > instruction set or do i have to define my own query language?
>
> I don't know what kind of database you want to implement.

I want my database to be a storage system which is neither object oriented nor relational. It just should offer tables with primary and secondary indices. So you could say it's pretty close to a relational database (just without primary and secondary keys and all the features and constraints that come with them). I think sometimes it would be useful if you (as a programmer) could just link a static library to your application and voila, you have database functionality. And you don't have to install another database runtime or ODBC-Driver or whatever...

Chris Received on Thu Jan 11 2001 - 18:07:16 CET