PoX Flash Memory: Fudan University Develops World’s Fastest Flash Memory to Power Future AI Systems

With potential applications in AI and big data processing, PoX memory could transform computing by enabling faster, more efficient storage solutions.

Introduction

A research team at Fudan University has developed the world’s fastest flash memory, called PoX Flash Memory, capable of reaching read/write speeds of 400 picoseconds.

This innovation could reshape how data is stored and processed in next-generation AI and high-performance computing systems.

Led by Professors Zhou Peng and Liu Chunsen, the team introduced a novel memory mechanism called “super-injection” based on quasi-2D materials. This breakthrough surpasses the speed limitations of traditional flash memory and even outperforms SRAM, the fastest volatile memory at the same technology node.

/techovedas.com/ascend-910d-huawei-targets-most-powerful-ai-processor-taking-aim-at-nvidia/

Quick Overview: PoX Flash Memory at a Glance

Record-Breaking Speed: Achieves 400 picoseconds per operation, enabling 2.5 billion operations per second.

Super-Injection Mechanism: Eliminates traditional delays by accelerating electrons instantly into storage layers.

Based on 2D Materials: Uses Dirac band structure and ballistic transport to enhance charge mobility.

AI & Big Data Ready: Ideal for processing massive datasets in real time.

Commercial Roadmap: Plans to scale to megabit integration and license the tech in 3–5 years.

Follow us on Linkedin for everything around Semiconductors & AI

Breaking Through the Memory Speed Wall

Flash memory forms the backbone of digital storage, from USB drives to SSDs. But as AI models grow larger and data volumes increase, traditional memory speeds have become a bottleneck.

Until now, the strategy to increase flash speed relied on letting electrons “run up” in the channel to gain energy before reaching the storage layer. This method had a natural speed ceiling due to the electric field distribution in semiconductors.

The Fudan team tackled this at the theoretical level, creating a quasi-2D Poisson model that predicted the super-injection phenomenon. This allows electrons to accelerate immediately without a “warm-up,” and inject into the floating-gate storage layer beyond conventional limits.

techovedas.com/wuji-china-unveils-worlds-first-1nm-risc-v-chip-built-with-2d-materials/#google_vignette

How PoX Flash Memory Works

The device still uses a floating-gate transistor, the standard unit in flash memory. But instead of relying on long electron acceleration paths, the team uses two-dimensional materials that feature ballistic transport—meaning electrons travel without scattering.

By tuning the Gaussian length of the 2D channel, the researchers created an ideal path for charge to move quickly and consistently. This leap allowed the device to operate at sub-nanosecond speeds, far exceeding previous theoretical limits.

PoX vs Traditional Flash Memory

FeatureTraditional FlashPoX Flash Memory
Speed~10–100 nanoseconds400 picoseconds
Operations/secUp to 1 billion2.5 billion
Storage MechanismGradual electron injectionSuper-injection
Energy EfficiencyModerateMore efficient
Ideal ForConsumer electronicsAI, Cloud, Big Data

A Leap Forward for AI and Cloud Computing

With the explosion of generative AI and real-time data analysis, high-speed memory is now a vital component in computing architecture. PoX memory can allow AI models to train faster, reduce latency, and potentially merge memory and storage layers in PCs and data centers.

That means future computers may no longer need DRAM and SSD separately, simplifying design and cutting costs.

What’s Next?

The team plans to scale the device to tens of megabits over the next 3–5 years. Once large-scale integration is achieved, Fudan University aims to license the technology for industrial use, potentially positioning China as a global leader in memory innovation.

techovedas.com/10-billion-rmbchina-unveils-3-projects-accelerating-semiconductor-growth

Conclusion

The development of PoX Flash Memory by Fudan University signals a monumental step forward in semiconductor technology. Its unmatched speed and efficiency offer a glimpse into the future of computing, where data flows faster, AI grows smarter, and storage becomes seamless.

For more of such news and views choose Techovedas! Your semiconductor Guide and Mate!

Kumar Priyadarshi
Kumar Priyadarshi

Kumar Joined IISER Pune after qualifying IIT-JEE in 2012. In his 5th year, he travelled to Singapore for his master’s thesis which yielded a Research Paper in ACS Nano. Kumar Joined Global Foundries as a process Engineer in Singapore working at 40 nm Process node. Working as a scientist at IIT Bombay as Senior Scientist, Kumar Led the team which built India’s 1st Memory Chip with Semiconductor Lab (SCL).

Articles: 2856

For Semiconductor SAGA : Whether you’re a tech enthusiast, an industry insider, or just curious, this book breaks down complex concepts into simple, engaging terms that anyone can understand.The Semiconductor Saga is more than just educational—it’s downright thrilling!

For Chip Packaging : This Book is designed as an introductory guide tailored to policymakers, investors, companies, and students—key stakeholders who play a vital role in the growth and evolution of this fascinating field.