--- title: "Micron Readies Client Storage for AI" type: "News" locale: "zh-HK" url: "https://longbridge.com/zh-HK/news/281383944.md" description: "Micron Technology is launching its first PCIe Gen5 QLC SSD aimed at mainstream PCs and ultra-thin laptops, anticipating a growing demand for local AI capabilities due to privacy concerns. The 3610 SSD boasts impressive performance with read speeds up to 11,000 MB/s and write speeds of 9,300 MB/s, making it suitable for AI workloads. Micron emphasizes the importance of power efficiency and thermal management in its design, addressing the tight NAND supply amid rising AI data center demands. The company believes the shift to on-premise AI solutions is inevitable as businesses seek control over their data." datetime: "2026-04-01T14:06:08.000Z" locales: - [zh-CN](https://longbridge.com/zh-CN/news/281383944.md) - [en](https://longbridge.com/en/news/281383944.md) - [zh-HK](https://longbridge.com/zh-HK/news/281383944.md) --- > 支持的語言: [简体中文](https://longbridge.com/zh-CN/news/281383944.md) | [English](https://longbridge.com/en/news/281383944.md) # Micron Readies Client Storage for AI Micron Technology is prepping desktops for AI. Its latest QLC SSD targets mainstream PCs and ultra-thin laptops and, according to the company, is the industry’s first PCIe Gen5 QLC SSD for client computing. Christopher Moore, VP of marketing for Micron’s mobile and client business unit, said Micron foresees commercial client computing to increasingly require local AI capabilities primarily due to privacy and data security concerns. “The need for more horsepower in a laptop or a desktop is going to be there.” In a briefing with EE Times, Moore said AI is penetrating many different industries, and businesses don’t want intellectual property sitting in a cloud server that they can’t control. “On-premise is going to be a very, very big mover due to AI. It’s a natural next step.” The migration of AI to the desktop will extend to other consumer devices, Moore said, including smartphones, as users decide they don’t want personal photos going back and forth to the cloud when editing them or using AI to generate something new. Shawn Adams, senior product line manager at Micron, said the transition to more intensive AI workloads being done on end-user devices is inevitable. “We know that it’s coming,” he said. “We know there are applications. It’s just a matter of timing. The applications have got to catch up.” Micron claimed its 3610 SSD delivers up to 11,000 MB/s sequential read speeds and 9,300 MB/s sequential write speeds along with 1.5 million random read input/output operations per second (IOPS) random read and 1.6 million IOPS random write. The company said this level of performance makes it “AI-ready.” The performance capabilities of the 3610 SSD stem in part from improvements at the NAND level, along with the integration of Micron’s Adaptive Write Technology (AWT), which the company introduced with its 2600 SSD. AWT is a NAND‑management scheme for QLC SSDs. It keeps the physical NAND as QLC, but changes how bits are programmed so each cell can temporarily act like SLC (1 bit), TLC (3 bits), or full QLC (4 bits) depending on workload and free space. Controller firmware continuously determines the appropriate mode based on the amount of data being written, drive fill level, and drive state. Moore said power efficiency is another key consideration for the client SSD segment. He noted that the DRAM-less architecture of the 3610 SSD, combined with host memory buffer (HMB) and DEVSLP (device sleep) low-power states, improves performance per watt by 43% compared with Gen4 TLC. The 3610 SSD also addresses thermal management with precise controls for OEMs to support ultra-thin, fan-less designs, the company said. Adams said some of Micron’s tests include loading 20 billion parameters of AI models in under three seconds, which is notable because such a power-intensive workload would have previously required a TLC SSD with DRAM. “That’s a big deal. We’ve moved QLC into a mainstream space.” The introduction of Micron’s 3610 SSD, based on the company’s G9 NAND, comes at a time when NAND supply is tight due to high demand from AI data centers, Moore acknowledged. “There is just a lack of supply to satisfy all the demand in every single segment.” He said QLC is inherently 25% more efficient in terms of bits per wafer. “It takes fewer wafers to make the same amount of bits, which means there’s more supply that’s going to the industry.” Behind Micron’s New 256GB LPDRAM: A Culture of Co-Design With Customers Micron’s $200 Billion Plan to Reshore U.S. Production Faces Challenges Gary Hilson is a freelance writer and editor who has written thousands of words for print and pixel publications across North America. His areas of interest include software, enterprise and networking technology, research and education, sustainable transportation, and community news. His articles have been published by Network Computing, InformationWeek, Computing Canada, Computer Dealer News, Toronto Business Times, Strategy Magazine, and the Ottawa Citizen. Follow Gary on LinkedIn Gary Hilson is a freelance writer and editor who has written thousands of words for print and pixel publications across North America. His areas of interest include software, enterprise and networking technology, research and education, sustainable transportation, and community news. His articles have been published by Network Computing, InformationWeek, Computing Canada, Computer Dealer News, Toronto Business Times, Strategy Magazine, and the Ottawa Citizen. You must Register or Login to post a comment. This site uses Akismet to reduce spam. Learn how your comment data is processed. ### 相關股票 - [Invesco Semiconductors ETF (PSI.US)](https://longbridge.com/zh-HK/quote/PSI.US.md) - [Direxion Semicon Bull 3X (SOXL.US)](https://longbridge.com/zh-HK/quote/SOXL.US.md) - [iShares Semiconductor ETF (SOXX.US)](https://longbridge.com/zh-HK/quote/SOXX.US.md) - [VanEck Semiconductor ETF (SMH.US)](https://longbridge.com/zh-HK/quote/SMH.US.md) - [GraniteShares 2x Long MU Daily ETF (MULL.US)](https://longbridge.com/zh-HK/quote/MULL.US.md) - [SPDR S&P Semicon (XSD.US)](https://longbridge.com/zh-HK/quote/XSD.US.md) - [Invesco PHLX Semiconductor ETF (SOXQ.US)](https://longbridge.com/zh-HK/quote/SOXQ.US.md) - [Direxion Daily MU Bull 2X Shares (MUU.US)](https://longbridge.com/zh-HK/quote/MUU.US.md) - [Micron Tech (MU.US)](https://longbridge.com/zh-HK/quote/MU.US.md) - [First Trust Nasdaq Food & Semicon (FTXL.US)](https://longbridge.com/zh-HK/quote/FTXL.US.md) ## 相關資訊與研究 - [Micron's stock is seeing its biggest gain in a year, en route to a record market-cap boost](https://longbridge.com/zh-HK/news/281405397.md) - [Micron's stock falls further, but this analyst says bears are missing the point](https://longbridge.com/zh-HK/news/280645089.md) - [Who Is TipRanks' 5-Star Analyst on SanDisk Stock (SNDK)?](https://longbridge.com/zh-HK/news/281177294.md) - [SanDisk Stock (SNDK) Tops S&P 500 in Q1 2026 – Is It Too late to Buy?](https://longbridge.com/zh-HK/news/281374457.md) - [Micron Technology Prices Tender Offers for Senior Notes](https://longbridge.com/zh-HK/news/281251689.md)