Re: Question from someone who is not trained in computer sciences

On Thu, Sep 5, 2019 at 2:00 PM Judith Lacoste <jlacoste(at)miacellavie(dot)com> wrote:
> I think PostgreSQL is the solution for my needs, but I am not a programmer/coder.

I don't think that it's restricted to people that are computer
scientists. At least, I certainly hope it isn't. SQL was originally
supposed to be something that is usable by domain experts/analysts,
rather than by computer people (that was at a time when the divide was
far larger than it is today).

> I plan to install the database on a server in the office. Me and my four colleagues will occasionally connect to this database when we are working in other locations (usually hospitals or universities). In such remote locations, we often do not have internet/network, yet we still need to access the database. Currently, we use a system where a copy of the database lives on each of our laptops. We can access all the information in the database despite being offline. This local copy of the database is synchronized with the server once network becomes available again.
>
> My question is whether or not such set up is possible with PostgreSQL?

Since you're a biologist, you may like to play around with the Mouse
Genome database using PostgreSQL:

http://www.informatics.jax.org/downloads/database_backups/

Any supported version of PostgreSQL will work. You'll need to use
pg_restore to restore the databases. Something like this will do it:

pg_restore -d mgd /path/to/mgd.postgres.dump

(I'm not sure what operating system you'll use -- something similar to
this invocation ought to work on Windows through cmd.exe, though.)

From there, you can play around with the database using a GUI tool
such as pgAdmin. I sometimes use this database to test certain things,
since it's the only example of a living, breathing PostgreSQL database
that you can just download that I am aware of. Its schema is probably
not an exemplar of good design, but it does seem reasonably well
thought out. I'm not a domain expert, though, so I can't really be
sure how good it is.

The nice thing about this approach is that you can figure it out using
a "top down" approach, by first understanding how the database is used
in practical terms, and then filling in the details of how the
application that it backs uses the database.

Last I checked, restoring this database will take about 30GB of disk
space on top of the dump file itself.

--
Peter Geoghegan