Disaggregating System Architectures for Future HPC and AI Workloads
An explosion in demand for more compute capacity for high-performance computing (HPC) and artificial intelligence (AI) workloads is about to usher in a new era of disaggregated system architectures – with an emphasis on increased flexibility, efficiency, speed, and reduced costs. This Technology Outlook summarizes a recent panel discussion among industry thought-leaders from Microsoft, Intel, Lawrence Livermore National Lab, Lawrence Berkeley National Labs and Ayar Labs. Entitled Disaggregating System Architectures for Future HPC and AI Workloads, the paper shares more about disaggregated systems and the drivers behind them, as well as what it will take to make them a reality for the future of HPC and AI.