Odroid XU4q
A perfect thin-client

Its already been a couple of weeks since my last post, I had hoped to post more often in the last couple of days but putting the fine details for my SQL database project in place has taken a lot more of my free time than I hoped + I don’t want to post unless my post have tested and personally development details !

I do feel in a great position now however to again start to post here, posts that include some interesting tested results of system builds and software installations and configurations.

Most of my time in the last couple of weeks has been spent working one putting in place a networked group of small single board computers including – Asus Tinker boards an Odroid XU4q and a raspberry pi SQL-database server.

I have been working hard to establish both the best available hardware to use and the best operating systems.

These are the system configurations I have now selected

1.. A Raspberry Pi 3b+ installed in an (element 14) PI-Desktop case, with Ubuntu Mate 16.04-LTS and Postgresql 9.5 (SQL database server) installed. This system has been configured to allow for remote client connections.

2.. As client systems to the above, I have installed both a Asus Tinker board with Armbian OS ( based on Ubuntu 18.04-LTS) and an Odroid XU4q with Ubuntu 18.04-LTS.

3.. An Asus Tinker board with Android 9 installed- I want to test this operating system to see just how much SQL client software is available along with testing just how much Python programming using psycopg2 and Tkinter modules (detailed below!) can be used on an android system?, it could be that I need to uses JAVA on this system for software development ?

As said – as software selection, I have installed PostgresQL 9.5 as the SQL server software on a Raspberry Pi 3B+, I feel that despite the lower spec of the Raspberry PI to the other boards I have used, this system should be more stable (running at lower power ratings and thus temperatures), an SQL server will run without a screen and thus the need for any great graphics performance.

As for the client systems and software, The extra power provided by the Asus Tinker boards and Odroid Xu4q will provide more than enough processor and graphics power to run all needed database client software and development processes.

As a last note : The other two areas worked on in the last two week relate to Python development and the first review of both Kexi an SQL client application.

With Python I have now worked with both the pyscopg2 and the tkinter modules for both python 2 and 3.

Psycopy2, is the most popular python module used with the python development language used in order to open a connection to an SQL-Server and then provide all the required API commands used as an interfaces, to perform the full set of processes for SQL programming.

Tkinter, this modules added to python a full set of commands and API’s needed to create even the both powerful of modern GUI applications.

These modules combined provide for all the missing links in an up-to date database system, the ability for python to perform all the available SQL operations and then to present data and retrieve data from the user in the form of a GUI is indeed completing a puzzle that many find in-front of themselves when they first start coding.

There is a Question however that could be asked and that is – if with very powerful database application such as (Libreoffice-base, MS-Access and Kexi), do you actually need to write python code?

I hope to find the answer to this question over the coming weeks, just for starters however , you have to remember that many application allow you to add python scripts as additions, this would allow you to access the data-set you are working from your SQL-server and then update this data or store results from the same application back to the same database server data-sets.

Another good reason to use Python, would be in an environment where data is both being stored and retrieved from your SQL server automatically, such as a sensor based AI system on a factor floor !