Longsys unveils Edge AI storage fusion strategy at FMS 2026

scanx
Reviewed by
ScanX News Team
Key Highlights

Longsys (301308.SZ) will present its Edge AI Storage Fusion strategy at FMS 2026 in California, August 4-6, 2026. The company will showcase integrated solutions for AI PCs, mobile devices, and intelligent hosts. Chief Scientist Dr. Chen Jian will explain the Storage Foundry model, covering chip design to manufacturing.

powered bylight_fuzz_icon
46714672

*this image is generated using AI for illustrative purposes only.

Longsys (301308.SZ) will unveil its "Edge AI Storage Fusion" strategy at the Flash Memory Summit (FMS) 2026, taking place in California, USA, from August 4 to 6, 2026. The Shenzhen-based semiconductor memory enterprise aims to demonstrate how its redesigned storage architecture bridges edge AI computing and storage, addressing industry challenges such as insufficient compute adaptation and high retrofit costs. This move signals Longsys's effort to transition the sector from technical exploration to commercial scale by redefining storage as a computing support hub rather than a passive data medium.

The presentation will focus on three core edge AI scenarios: AI Box intelligent hosts, AI PC terminals, and AI Mobile embedded devices. Longsys intends to showcase integrated hardware-software solutions that leverage proprietary technology and supply chain synergy. The company claims these solutions deliver multidimensional upgrades in performance, efficiency, and cost across diverse terminal scenarios.

Technological and Value Fusion

The "Edge AI Storage Fusion" theme encompasses two primary layers of innovation. The first is technological fusion, which involves integrating chips, storage, memory, and intelligent scheduling algorithms. This approach aims to break down traditional barriers between storage and memory to achieve true storage-compute synergy.

The second layer is value fusion, which seeks to redefine the role of storage in edge AI ecosystems. By aggregating hardware performance, software efficiency, and ecosystem adaptability, Longsys aims to enable lower-cost, higher-performance, and more scalable edge AI deployment. This philosophy underpins the company's efficient and deployable solutions for terminal AI applications.

Storage Foundry Model

Chief Scientist Dr. Chen Jian will present a session titled "Intelligence at the Edge: Powered by the Storage Foundry Model." This model represents an innovative business approach that moves beyond isolated hardware upgrades. Instead, it spans the full value chain, including:

Value Chain Stage Description
Chip Design Customized design processes
Hardware & Firmware Integrated system development
Packaging & Testing Industrial design and quality assurance
Materials & Manufacturing End-to-end production capabilities

Dr. Chen will detail how deep cross-functional collaboration enables full-stack, efficient customization from design through delivery. Longsys has deeply invested in this model to support global partners in exploring new opportunities at the intersection of edge AI and storage.

What the Numbers Show

While specific financial metrics were not disclosed in the announcement, the strategic shift toward a full-stack Storage Foundry model indicates a move toward higher value-added services. By controlling the entire process from chip design to manufacturing, Longsys aims to reduce dependency on external components and improve margin structures through proprietary technology integration. This vertical integration strategy positions the company to capture more value within the growing edge AI hardware market.

How might Longsys's vertical integration strategy impact its gross margins compared to traditional storage manufacturers in the 2026-2027 fiscal periods?

Which major global semiconductor or AI hardware partners are expected to adopt Longsys's 'Storage Foundry Model' for their next-generation edge devices?

What specific regulatory or supply chain risks could hinder the commercial scaling of Longsys's proprietary chip design and manufacturing capabilities?

like16
dislike

Longsys achieves one million monthly mSSD production capacity

scanx
Reviewed by
Anirudha BScanX News Team
Key Highlights

Longsys achieved a stable monthly production capacity of one million mSSD units in June 2026, utilizing SiP integration technology to manufacture PCIe Gen4 and Gen5 solutions. The company is collaborating with Lenovo and ASUS to deploy these storage solutions for edge AI devices, AI PCs, and intelligent terminals.

powered bylight_fuzz_icon
44543694

*this image is generated using AI for illustrative purposes only.

Longsys announced the establishment of a stable monthly production and delivery capacity of one million mSSD (Micro SSD) units in June 2026. This milestone marks a significant advancement in the company's high-speed storage solutions, designed to meet growing market demand and support future expansion. The production ramp-up follows the initial rollout of mSSD samples at the company's Suzhou packaging and testing manufacturing base in October 2025.

The new capacity leverages System-in-Package (SiP) integration and packaging technology to innovate traditional storage form factors. The performance, quality, and functionality of these products have been validated through real-world customer applications. Longsys is currently working with leading industry customers, including Lenovo and ASUS, to integrate these solutions.

Product Specifications and Technology

Longsys currently offers both PCIe Gen4 and PCIe Gen5 mSSD solutions. These products are designed with customized thermal architectures to deliver sustained high performance within a compact form factor suitable for edge AI devices. The PCIe Gen5 version incorporates a vapor chamber (VC) cooling solution to support higher bandwidth, larger capacities, and improved sustained performance under intensive workloads.

The mSSD adopts a SiP architecture that integrates the controller, NAND flash, power management IC, and passive components into a single package. Compared to conventional PCB-based SSD designs, this architecture simplifies manufacturing, improves space efficiency, enhances product reliability, and shortens production cycles.

Feature Specification
Architecture System-in-Package (SiP)
Interface PCIe Gen4, PCIe Gen5
Cooling (Gen5) Vapor Chamber (VC)
Production Capacity 1 million units/month

Strategic Developments

The SiP integration solution is compatible with multiple generations of high-speed PCIe specifications, including Gen4, Gen5, and the upcoming Gen6. It delivers product development capabilities covering consumer-grade, industrial-grade, and automotive-grade reliability standards.

To strengthen its edge AI storage portfolio, Longsys introduced its proprietary SPU (Storage Processing Unit), iSA (Intelligent Storage Agent), and HLCache technologies in March 2026. These technologies create an integrated hardware-software architecture for next-generation intelligent storage.

Looking ahead, Longsys plans to expand its mSSD platform across AI PCs, AI Boxes, and other edge AI devices while advancing the PCIe Gen5 ecosystem. The company aims to accelerate the large-scale deployment of edge AI storage solutions by combining SiP integration with hardware-software innovation.

How will Longsys manage the supply chain logistics to sustain the one million unit monthly capacity given the volatility in NAND flash pricing?

What is the projected cost per unit of the SiP-based mSSD compared to traditional PCB-based SSDs, and how will this impact pricing in the edge AI market?

When does Longsys anticipate the commercial availability of PCIe Gen6 solutions based on their current SiP architecture roadmap?

like20
dislike