Niv-AI exits stealth to wring more power performance out of GPUs
Niv-AI has emerged from stealth with $12 million to use new sensors and AI tools to manage GPU power usage efficiently in data centers.Summary
Electricity management is a critical bottleneck for AI, with data centers often throttling GPU usage by up to 30% due to power management difficulties caused by millisecond-scale power demand surges during computation tasks. Niv-AI, a Tel Aviv-based startup, has exited stealth with $12 million in seed funding to address this inefficiency. The company is deploying rack-level sensors to measure GPU power usage at millisecond precision to understand specific power profiles. Niv-AI plans to use this data to build an AI model, acting as a "copilot" to predict and synchronize power loads, ultimately creating a missing "intelligence layer" between data centers and the electrical grid. This aims to help data centers utilize more of the power they already pay for while creating more responsible power profiles for the grid.
(Source：TechCrunch)