Multi GPU Training

Multi GPU Training

As Moor's Law is working its way into retirement one can not just rely on computations to become faster as the underlying hardware improves. There are still plenty of ways to cram more FLOP into the second but most of them rely on running more of them in parallel as opposed to in sequence, but quicker. Think of this as a guide to the basics. Read more

The Damage a GPU Can Do

The Damage a GPU Can Do

The energy usage of machine learning applications can be immens. This often goes hand in hand with catastrophic greenhouse gas emissions. This is how bad the situation really is and why classifying cats in Iceland is the sensible thing to do. Read more