HW-based Methods to Dynamically Throttle AI Performance (Princeton University)

Researchers from Princeton University published a technical paper titled “Hardware Mechanisms to Dynamically Throttle AI Performance.” Abstract Excerpt: “In this paper, we introduce a set of microarchitecture knobs which dynamically control the available hardware resources to limit AI performance at runtime. We evaluate candidate knobs spanning the GPU memory subsystem, across capacity, bandwidth, latency and frequency dimensions, and narrow down to four strong candidates: L2 size, L2 latency, L