SQL PDF course

objectives

At the end of this course, the participant will have acquired SQL Server basic knowledge to create databases, tables, views, procedures, functions and triggers. It will be able to enter, modify and delete data from tables and define constraints on tables. He will also know how to back up and restore a database.

Pedagogical approach

Overview of functions and commands
Demonstration of computer commands
Computer exercises supervised by the trainer

Talking about SQL nowadays as a novelty, would be a challenge ... However, should not we see in this indestructible language, an attempt a bit late, but desirable, through the various middleware available, standardization of a data query mode?
It is clear that even the object databases and the web are going to SQL. The weight of the past without doubt ...

But what to do? Squeezer SQL or live with it?

More than twenty years ago, COBOL was, it was said, assured of certain and short-term death. Today the deceased is still panting although tired. Will it be the same for SQL? All suggests!

SQL a language?

Is SQL a real language?

If one must accept this word in the computer sense of the term, it seems difficult to say yes as SQL is far from the structure and possibilities of a common programming language. Variable point, procedure or function point ... However, it is important to form sentences that will be compiled in order to execute treatments.
Although SQL has developed over time, such as the ability to parameterize requests, there are significant gaps that are hindering its penetration. For example, SQL can not be recursive while this mode of execution is necessary to solve a large number of problems, especially those to treat trees or graphs.

In fact, SQL is a "declarative" language. We specify what we want to get or do, and it's the machine that decides how to do it. For some, SQL is seen as a last resort while others prefer to avoid it or delay at most the inevitable moment when its appearance will be unavoidable.

The fundamental difference between common languages like C or Pascal, which are procedural languages, lies in the fact that with these, you indicate all the instructions necessary to treat a problem. This way, you keep a certain control over the execution cycle and the sequence of the various tasks of a program is perfectly known to you. On the other hand, in SQL you do not have to have any idea how the machine executes your request, or even in what order it breaks down the processing of the request into different tasks, or even even how it synchronizes them.

During an extraction concerning several tables, the relational engine could perfectly launch several parallel processes to extract the data of each of the tables, then to make the joins of the various resulting tables according to the order of termination of the extractions ...
This is why relational drivers include logical and sometimes statistical optimization modules.

Thus a logical optimizer will prefer to deal first with the clauses excluding a maximum of data from the response, while a statistical optimizer will start processing the tables of lower volume.