1. Researchers at the University of Minnesota have developed a magnetic tunnel junction-based device that could reduce AI energy consumption by over 1,000 times. 2. The device utilizes computational random-access memory (CRAM) to perform computations within memory cells, eliminating the need for data transfers between logic and memory. 3. This innovation aims to enhance energy efficiency and reduce costs in AI computing applications.
Related Articles
- With AI, Digital Twins, and Smart Algorithms for a Resource-Efficient Energy Economy4 months ago
- Harnessing Biology for Sustainable Energy Production6 months ago
- AI datacentres will need 160% more electricity by 20277 months ago
- Microsoft claims timber-built datacenters can reduce its carbon footprint by up to 65%7 months ago
- Alexander von Humboldt Professorship for Luisa Petti1 day ago
- AI drone beats human champions for the first time at Abu Dhabi racing event – new deep neural network sends control commands directly to motors in significant leap1 day ago
- TUM Ranks 7th in Educating 'Digital Leaders'2 days ago
- Broadcom shipping Tomahawk 63 days ago
- Smart 4K Camera For Work And Learning3 days ago
- RUSI report highlights UK ‘defence dividend’ for local areas3 days ago