The CRAM device, developed by scientists over the past 20 years, is set to revolutionize artificial intelligence (AI) computing by drastically reducing energy consumption.
Researchers at the University of Minnesota have achieved a breakthrough with the CRAM (Computational Random Access Memory) device, which has the potential to cut energy consumption for AI applications by a factor of 1000.
CRAM: Artificial intelligence will reduce energy consumption by 1000 times
In traditional AI methods, data constantly moves back and forth between processing units and memory, consuming significant amounts of energy. However, CRAM addresses this issue by eliminating these energy-intensive transfers. How does it achieve this? CRAM processes data directly within memory cells, effectively removing the need for data to travel between processing and memory units. This approach streamlines the process by performing data processing directly inside the memory.
This new device is built on magnetic tunnel junctions (MTJ) and nanostructured devices, which offer several advantages over traditional transistor-based chips, including higher speed, lower energy consumption, and greater resistance to harsh environments.
It is estimated that a CRAM-based machine learning accelerator could save up to 2,500 times more energy compared to traditional methods. This remarkable energy efficiency has the potential to significantly impact the future of AI. The research team plans to collaborate with semiconductor industry leaders to further develop CRAM technology and enhance AI capabilities.
Yang Lv, the lead author of the study, states, “This work represents the first experimental application of CRAM where data can be processed without leaving memory.” The team has spent over 20 years developing this technology and believes it is now ready for real-world applications.
Traditional computer architectures consume significant amounts of energy during data transfers. In contrast, CRAM revolutionizes this process by processing data directly within the memory cells, resulting in remarkable energy savings. It offers substantial improvements in both speed and efficiency. You can find the details of the article here.
What are your thoughts? Feel free to share your views in the comments section below.
You may also like this content
- Meta Building World’s Fastest AI Supercomputer for Metaverse
- Artificial Intelligence Will Make Decisions Instead Of People
- Bill Gates: Artificial Intelligence Over Web3 and Metaverse
-
-
-
-
-
-
-
-
-
-
