Wednesday, November 21, 2018

Which Flavor of Linux

This week I have been working from home and been doing some playing around with machine learning (ML) in Python. My home Linux box is using an old version of Fedora Core while my machine at work uses Ubuntu Linux. I am having problems getting Python updates to work correctly because some of my system libraries are so old. It is time to update my Linux and I am wondering which version to select.

I started using Red Hat Linux a long time ago. I don't feel the need to pay for Linux as I don't use it as a production server and so I have been running Fedora Core, which is based on Red Hat. If I want to continue using a Red Hat flavor of Linux, I could also install CentOS. So should I stay with Fedora Core or change to CentOS?

As mentioned, my machine at work uses Ubuntu which is based on the Debian distribution of Linux. I like it and it stays updated fairly easily. There are several other options as well but most of the world has settled on Debian or Red Hat. Suse used to be an option but Novell did a pretty good job of killing it in the early 2000's.

So what is the difference between them all? Mostly it has to do with system administration. PC users don't really want to do a lot of system administration work but they do like to install new versions of software. Red Hat uses "yum" to install software while Debian uses "apt-get". The programs are very similar. I have never had a problem finding updates for either flavor of Linux.

One problem I have noticed with Fedora Core is they are constantly changing versions. You may be on version 22 only to have version 23 come out a few months later. Updating your operating system can be a bit of a time sync as you have to make backup copies of your files and then spend hours making sure everything got moved to the same version without any problems. Ubuntu doesn't have that problem and so I think I will make that change over the Christmas break when I have more time. Until then I will have to remote login to my machine at work when running through my ML exercises. You may disagree and I would love to hear why. Feel free to send me a comment.

No comments:

Post a Comment