Chinese Scientists Achieve 10,000x Speed Breakthrough in Flash Memory Technology

· 1 min read

article picture

Chinese researchers at Fudan University have achieved a groundbreaking advancement in computer memory technology, developing a new type of flash storage that operates 10,000 times faster than current solutions.

The novel memory device, dubbed "PoX," can write data in just 400 picoseconds - equivalent to performing 25 billion operations per second. This breakthrough pushes non-volatile memory into speed ranges previously only possible with volatile memory types like SRAM and DRAM.

Led by Professor Zhou Peng at the State Key Laboratory of Integrated Chips and Systems, the research team redesigned traditional flash architecture by replacing silicon channels with graphene - an ultra-thin material just one atom thick. This allows for dramatically faster data transfer while maintaining the ability to retain information without power.

"Using AI-driven process optimization, we drove non-volatile memory to its theoretical limit," explained Professor Zhou to Xinhua News Agency. The speed improvement is dramatic - while current flash memory manages around 1,000 writes per second, PoX can perform 1 billion operations in the blink of an eye.

The technology could revolutionize artificial intelligence systems, which require massive amounts of high-speed data transfer. Current AI hardware is often bottlenecked by slow memory speeds rather than processing power. PoX memory could eliminate the need for separate high-speed cache memory in AI chips, reducing both size and power consumption.

The development also has strategic importance for China's semiconductor industry. While the research team has not yet revealed details about production yields or long-term reliability, the technology appears compatible with existing manufacturing processes for 2D materials.

The Fudan University team is now working on scaling up the technology for practical applications. If successful, PoX could enable a new generation of ultra-fast, energy-efficient computers and mobile devices while meeting the growing storage demands of advanced AI systems.