PMBus changes improve power efficiency

摘要

Uptime and energy efficiency are now the two most important factors in the design of infrastructure equipment such as computer servers and network switches and gateways. The key to both is understanding in real time, what is happening inside the system. Today's infrastructure equipment comprises multiple blades, often with multiple system-on-chip (SoC) and field-programmable gate array (FPGA) devices on each. The distributed power architecture common to these systems allows fine-grained power control for each, often with support for hot swap to allow failed components to be switched out and replaced without affecting the rest of the system. Ensuring that the power delivered to each blade and processor is within tolerance and at the appropriate levels demands the ability for the system controller to understand what is going on. A key component of that architecture is the Power Management Bus (PMBus).