Micron Technology officially releases brand-new client-oriented solid state drives, focusing on empowering local AI computing on personal terminal devices. The newly launched product is the industry’s first PCIe Gen5 QLC SSD specially designed for mainstream desktop computers and ultra-thin laptops.

Figure: Micron’s PCIe Gen5 QLC client SSD built with G9 NAND (Source: Micron Technology Official Release)

Industry insiders from Micron stated that more enterprises and individual users tend to deploy AI computing locally rather than relying on cloud servers, mainly driven by data security, privacy protection and independent intellectual property management demands. Local offline AI operation has become an inevitable development trend in client computing market.

This local AI trend is not limited to PC terminals, and will gradually spread to smart phones and other consumer electronic devices. Users are increasingly inclined to finish AI image editing, content generation and other tasks on local devices to avoid frequent data transmission and privacy leakage risks.

Named Micron 3610 SSD, this new storage product delivers outstanding hardware performance. It features up to 11,000 MB/s sequential read speed and 9,300 MB/s sequential write speed, with powerful random access capability to fully meet high-load edge AI operating requirements.

The SSD adopts Micron self-developed G9 NAND flash particles and is equipped with mature Adaptive Write Technology. The intelligent adjustment mechanism enables QLC particles to switch flexibly among SLC, TLC and QLC operating modes according to actual workload, greatly balancing performance, endurance and service life.

In terms of energy consumption and layout design, Micron 3610 SSD uses DRAM-less architecture, matched with HMB and DEVSLP low-power technologies. Its power efficiency is increased by 43% compared with traditional Gen4 TLC SSD. Meanwhile, it supports precise thermal control, perfectly fitting fan-less ultra-thin device design.

Actual verification shows that the SSD can complete loading of 20 billion-parameter AI models within 3 seconds, realizing high-speed startup and running of large local AI models. It successfully promotes high-performance QLC storage to enter mainstream client AI application scenarios.

Facing the current tight global NAND supply caused by explosive demand of data center AI business, QLC storage owns inherent cost and output advantages. Higher wafer bit utilization can effectively ease market supply pressure and accelerate the popularization of edge AI client devices.