skip to content

Department of Computer Science and Technology

 

Power consumption is increasingly important in High Performance Computing (HPC). There is a desire to be greener, but no clear path to achieve that in a resource and energy-hungry industry. The need for HPC has grown as more industries turn to simulation to make efficiency gains, but nothing in HPC compares with the explosion of compute needed for AI and crypto currencies. Power consumption is quickly becoming a limiting factor in the delivery of compute for machine learning and its applications. We don't yet have self-driving cars for example, and we will not until we have solved the power problem. More critically for the planet, the move to green energy and vehicle electrification is not happening quickly enough. We need to look to the next wave of computing to solve this problem. Is that AI or do we need to look out further towards quantum computing? In this talk I will look at the application of large scale compute on our ability to deliver the technology changes that we need to save the planet and I will look at the role power consumption plays in that.