Week 2

This week was spent setting up the AWS GPU instances that I'll be using. Typically you can use a local VM of HPCC systems so that you can have a small development cluster to test code with without the expense and overhead of having your own cloud based cluster. However, one of the challenges to using GPUs is that you can't use your GPUs on a local virtual machine. i.e. even if you have a GPU on your local computer the current VM solutions don't allow you to use them. This means to have an HPCC System on a computer that has a reachable GPU, you have to have it installed as a "bare metal" setup. This just means you install it on Linux as its own computer, not in a VM. You could of course have two physical computers at your desk, one for coding and one for GPU/HPCC, but I will be using AWS machines.