Our history with databases part two

[Andrea Guidi]

#113054

Our history with databases part two
LinkedIn newsletter February 15, 2023

In this second episode I continue with the intertwining of our history as Softpi and the world of databases. And here I finish! I promise.

Data replication is another aspect that has made us passionate. Real-time replication of all or part of the database, according to certain criteria, on heterogeneous systems, for the most diverse needs, such as having a separate system dedicated to Business Intelligence.

From Sybase’s Replication Server, which requires programming, to the more intuitive, but no less powerful, systems designed to make complexity simple, such as Syniti Replication. With or without data transformation while moving it. From dual systems, to star systems or database chains that pass data like the ball on a basketball court, until you make a basket.

At one point we dedicated a year to SAP Hana, study and work, supported a leading Italian customer and, from a completely different front, tried the Hana interface for integration with Omnis Studio.

I have spoken of experiences relating above all to systemic aspects, but the faces of databases have a multiple nature. There is conceptual and logical modeling, model maintenance and evolution, data selection and manipulation, standard SQL and its dialects, store procedures, multiple database integration, multi-level architectures, and more. the cloud and, continually trespassing into new territory, data lakes, data meshes, and so on. In various capacities we have cooperated and used Silverrun, I remember my friend Axel Troike when he came to present it at a meeting with a hundred of our customers in a pleasant farm in the Apennines, PowerDesigner, AimBetter, and here we are dealing with a recent acquaintance , with the great Yehuda Lasri.

We all know how DataBase Management Systems (DBMS) have characteristics such as: independence of data from the application, confidentiality in accessing data, management of physical integrity of data, management of logical integrity of data, security and optimization in use of data. To make the most of them we are helped by proprietary administration interfaces and their extensions and by tools such as DBArtisan.

Relational DBMSs, Relational DataBase Management Systems (RDBMS) were the first to be based on a previously formalized data model. This model was developed in 1970 by Edward Codd at the IBM San Josè Research Laboratory, with an approach based on the mathematical theory of the relationships between sets. It is independent of technical specifications and considerations relating to ease of implementation and efficiency of performance.

Given finite sets of values D1, D2, …, Dn, a relation R between these n sets is a set of n-tuples <d1, d2, …, dn>, where d1 belongs to D1, d2 belongs to D2, … dn belongs to Dn; D1, D2, …, Dn are called domains, their names are called attributes, n is called degree of relation R. I quote from McGraw-Hill’s 2004 SQL Guide book that we had the adventure of writing.

Now yes, I keep my promise and end up.

[Autore]

Post