Flash memory

Flash memory emerged from Fujio Masuoka's obsession with non-volatile storage. At Toshiba in the late 1970s, he was frustrated by EEPROM's limitations: it could be electrically erased, but only byte by byte, which made it slow and expensive for applications requiring bulk data storage. In 1980, Masuoka recruited four engineers to a semi-secret project aimed at creating memory that could be erased all at once—"in a flash."

The adjacent possible had aligned through two decades of floating-gate research. Intel's EPROM (1971) showed that electrons could be trapped on a floating gate to store data, but erasure required UV light. EEPROM (1978) added electrical erasure but needed two transistors per cell, making it expensive. Masuoka's breakthrough was reducing each cell to a single transistor while enabling block erasure—a seemingly small architectural change with enormous cost implications.

The name "flash" came from Masuoka's colleague Shōji Ariizumi, who noted that the bulk erasure reminded him of a camera flash. Masuoka and Hisakazu Iizuka filed the first patent in 1980. He presented NOR flash at a 1984 conference, then NAND flash at the 1987 IEEE International Electron Devices Meeting in San Francisco. Toshiba launched NAND flash commercially in 1987; Intel introduced the first commercial NOR flash chip in 1988.

NOR and NAND represented different trade-offs. NOR flash allowed random access like RAM, making it suitable for code storage and execution. NAND flash sacrificed random access for density and cost efficiency, making it ideal for data storage. This architectural split—one for firmware, one for files—would define the market for decades.

The cascade from flash memory eliminated mechanical storage in portable devices. Memory cards replaced film in digital cameras. USB drives replaced floppy disks. MP3 players became practical when flash costs dropped below spinning-disk alternatives. The smartphone revolution depended on NAND flash for apps, photos, and music. And solid-state drives eventually displaced hard disks in laptops and servers, eliminating the last spinning platters from many computing environments.

Masuoka never received the recognition or compensation his invention merited. He left Toshiba in 1994, later suing the company for inadequate invention compensation—a common grievance among Japanese engineers whose employers owned their patents. The lawsuit settled in 2006. By then, flash memory had become a $20 billion industry, with Toshiba and Samsung as dominant manufacturers.

By 2026, flash memory is ubiquitous—in phones, cars, data centers, and IoT devices. NAND density improvements continue, stacking cells vertically in 3D architectures. The technology Masuoka developed in secret at Toshiba became the storage substrate of the mobile era, proving that the difference between two transistors per cell and one transistor per cell was the difference between niche and foundational.