setting the stage – prepare your environment
This is what you want: “Get me a working version of a DFT code on my computer. Now.” And that’s what you will get. However fair your request is, though, it’s not an easy one to fulfill, for the following reasons:
- We don’t know which computer you are using. It may have Windows, macOS or one of the many Linux flavours as an operating system.
- We don’t know what’s your background on installing new software. Do you know how to compile source code and link it to the relevant libraries, or does this sentence sounds as guru-talk to you?
- We don’t know which DFT code you want to install.
To get you up and running quick and painlessly, we will use in this course your preferred DFT code via a virtual machine.
A virtual machine mimics an operating system within another operating system. Say you have a Windows computer, but want to run Ubuntu Linux. After having installed the proper virtual machine, you will have an Ubuntu Linux box on your Windows pc, where you can work exactly as if you were using a computer with Ubuntu Linux as its operating system.
Once you work inside the virtual machine, you have exactly the same environment as everyone else in the course. If you run into an error, anyone will be able to reproduce that error by doing exactly the same steps you did. “It must be due to a peculiarity of your computer” is a kind of argument you should not hear here.
No, a virtual machine is only relevant for training purposes. All resources your computer spends to the hard work of simulating a different operating system, cannot be used to do the DFT calculations. A DFT code will always run slower on a virtual machine than on a native operating system. That does not harm for the not very demanding calculations in this course. If you decide later to use DFT for real research work, you will most likely turn to a computing center anyway. But let’s not bother you with that step in the training phase.
Virtual machines are created and tested by the developers of a particular DFT code, and are contributed to this course by them. You can make the practical tasks in this course with the DFT code of your choice. The supported codes are listed here — follow the link of your preferred code for installation instructions:
- virtual machine for Quantum Espresso (installation instructions)
- <list will become longer in the near future>
All instructions for the practical exercises in this course will be explained step by step for any of the DFT codes listed above.
Note for users of an Apple MacBook M1 : the M1 generation of Apple uses an ARM-chip that is not and will never be compatible with the VirtualBox virtualization software. You can still get access to the Quantum Mobile virtual machine via Docker, rather than via VirtualBox (allthough you’ll miss some graphical features). Short instructions how to do this can be found on the launch page of Quantum Mobile (section ‘Docker container’). An alternative can be to use an UTM virtual machine for Mac, as explained in this student guide.