Available Technologies

Technology

Modern flash memory systems predominantly prioritize long-term data integrity rather than maximizing energy efficiency. However, edge computing and Internet of Things (IoT) applications tend to run with limited power sources and rely heavily on approximate short-lived data rather than highly accurate long-term data. As these edge computing applications continue to emerge and grow, new opportunities arise for energy-efficient approximate storage systems.

Researchers at UAH have developed a technology called EXPRESS that uses a novel partial programming technique to improve energy efficiency of NAND flash operations. It has the ability to reduce energy consumption to less than 75% of the traditional operation while still maintaining over 99% accuracy. This technology reduces this disproportionate energy-accuracy tradeoff and achieves the optimal balance for many IoT applications for which low-energy operation is more important than long-term accuracy.

Even though NAND flash memory systems require less power than other similar storage solutions, they still account for a significant portion of the total energy consumed by many IoT and edge computing applications. This type of application—which includes wearable electronics, smartphones, self-driving cars, smart homes, smart home appliances, etc.—is expected to grow by over 200% globally from 2021 to 2026. This technology will only expand the ways in which consumers can use edge computing and IoT.