Latest news with #NANDflash
Forbes
7 days ago
- Business
- Forbes
Sandisk And SK Hynix Agreement Enables Fast NAND Flash To DRAM In HBM Packages
Several NAND flash manufacturers were discussing higher bandwidth flash technologies at the 2025 FMS in Santa Clara, CA, but the announcement by Sandisk and SK hynix that they would work together to create standards enabling enable high bandwidth flash in an HBM module would enable the next generation of AI training and inference with lower costs and reduced energy consumption. Sandisk announced it has signed a Memorandum of Understanding with SK hynix to work together to drive standardization of High Bandwidth Flash memory technology. HBF is a NAND flash-based solution contained in an HBM package. HBMs are widely used to support the immediate memory needs of GPUs for AI training and inference. HBMs have traditionally used dynamic random-access memory, DRAM, providing very fast access to data. An image of a flash based, HBF module architecture is shown below. HBF is a new technology designed to deliver breakthrough memory capacity and performance for the next generation of AI inferences and would supplement the traditional DRAM based HBM, providing a lower cost per capacity option to support the HBM DRAM. Through this collaboration with SK hynix, the companies aim to standardize the specification, define technology requirements, and explore the creation of a technology ecosystem for High Bandwidth Flash. As AI models grow larger and more complex, inference workloads demand both massive bandwidth and significantly greater memory capacity. Designed for AI inference workloads in large data centers, small enterprises and edge applications, HBF is targeted to offer comparable bandwidth to High Bandwidth Memory (HBM) while delivering up to 8-16x the capacity of HBM at a similar cost. In addition to providing more memory using NAND flash in HBM-like packages, NAND-flash is also a non-volatile memory while DRAM is a volatile memory. That means that DRAM requires regular refreshes of the data it contains and this consumes energy. By substituting some non-volatile memory for what otherwise would be volatile memory in GPUs and AI applications it may be possible to reduce the energy requirements for AI applications in data centers. This could open up more AI opportunities for AI applications in energy constrained data centers, thus democratizing the development of AI applications. It could also help to reduce the projected power requirements for hyperscale and other data centers to support AI development. Sandisk also announced the formation of a Technical Advisory Board to guide the development and strategy of its HBF memory technology. The board, consisting of industry experts and senior technical leaders from both within and outside of Sandisk, will provide strategic guidance, technical insight, market perspective, and shape a standards-driven ecosystem. Enabled by Sandisk's BiCS technology and proprietary CBA wafer bonding and developed over the past year with input from leading AI industry players, Sandisk's HBF technology was awarded 'Best of Show, Most Innovative Technology' at FMS: the Future of Memory and Storage 2025. Sandisk targets to deliver first samples of its HBF memory in the second half of calendar 2026 and expects samples of the first AI-inference devices with HBF to be available in early 2027. Sandisk and SK hynix announced that they would work together to create standards for high bandwidth NAND flash, HBF modules to supplement DRAM-based HBM and enabling the next generation of AI.
Yahoo
05-08-2025
- Business
- Yahoo
Phison and Supermicro Collaboration Delivers Unmatched Storage Density for AI and Hyperscale Workloads
SAN JOSE, August 05, 2025--(BUSINESS WIRE)--Phison Electronics (8299TT), a global leader in NAND flash controllers and storage solutions, is collaborating to showcase one of the most advanced server solutions in the market with Supermicro, a total IT solution provider for AI/ML, HPC, cloud, storage and 5G/edge, to address growing demand to support high-density workloads. Through this collaboration, customers using Supermicro's Petascale Storage Family will be able to leverage Phison's high-capacity 122.88 TB Pascari D205V SSD, featuring a unique E3.L form factor and Gen5 NVMe performance. The result is a purpose-built solution to redefine the possibilities of storage density, thermal efficiency, and scalability in enterprise infrastructure. As enterprises tackle infrastructure shifts to support data-intensive workloads such as AI and machine learning training, real-time analytics and cloud-scale storage, striking a balance between high-performance and scalable capacity is more critical than ever. Through this collaboration, Phison will deliver storage solutions to support Supermicro's Petascale Storage with unmatched capacity per slot, enabling users to reduce total rack space, lower operational costs and simplify infrastructure planning at scale, whether at the edge or in the data center. The E3.L form factor is part of the broader EDSFF family offered in the Pascari D-Series and is specifically engineered for high-density, high-performance environments and designed to be hot-pluggable and front-accessible. Compared to legacy U.2 and U.3 designs, E3.L offers a longer form factor which unlocks double the capacity compared to E3.S as well as improved airflow and thermal management, making it ideal for AI training clusters, hyperscale environments, and dense edge deployments where cooling and space are critical constraints. "This innovative collaboration with Supermicro sets a precedent to keep pace with the increasing storage demands of tomorrow," said Michael Wu, GM and President at Phison US. "Customers can expect their storage solutions to have built-in scalability and cost optimizing features from the drive to the rack architecture." The Pascari D205V and Supermicro solutions will be in Supermicro's display at FMS booth #846. To learn more visit the Pascari web page and Phison blog. About Phison Electronics Corporation Phison Electronics is a global leader in NAND flash controllers and storage solutions, powering more than one in every five SSDs shipped worldwide. Phison has grown into a multi-billion-dollar company with over 4,500 employees—70% of which are dedicated to R&D – and more than 2,000 patents. The company's innovations include aiDAPTIV+, an award-winning AI solution for affordable LLM training and inferencing on-premises, and Pascari, a portfolio of ultra-high-performance enterprise SSDs purpose-built for data-intensive workloads across AI, cloud, and hyperscale data centers. View source version on Contacts PHISON SpokespersonAntonio YuTEL:+886-37-586-896 #10019Mobile: +886-979-105-026Email: antonioyu@ PHISON Deputy SpokespersonKuo-Ting LuTEL: +886-37-586-896 #26022Mobile: +886-979-075-330Email: kuoting_lu@ PHISON and PASCARI enterprise product inquiries:sales@ Sales@ PHISON and PASCARI enterprise media inquiries:Lynn KellyLynn_kelly@ press_americas@
GSM Arena
22-05-2025
- Business
- GSM Arena
SK Hynix unveils 321-layer UFS 4.1 that is faster, thinner and more efficient than the previous gen
Memory maker SK hynix has announced the world's first 321-layer UFS 4.1 TLC NAND flash for smartphones. It's faster, more efficient and thinner to boot – perfect for the next generation of phones that will focus on slim builds and AI tools, says the company. Compared to the previous generation (which used a 238-layer design) from 2022, these new storage chips have 15% higher random read and 40% higher random write speeds. For sequential reads, they max out the interface at 4.3GB/s. On top of that, the NAND package is 0.85mm thick, down from 1mm. It doesn't sound like much, but every little bit helps if phones like the Galaxy S25 Edge become popular. Recent trends aside, the new 321-layer UFS 4.1 design is 7% more power efficient than the previous generation – less heat and more performance are always on trend. SK hynix says that the sequential read speed will improve on-device AI performance (since it will speed up loading the model to RAM), while the improved random performance will boost multitasking. The company will produce storage in two capacities – 512GB and 1TB. That's right, there isn't going to be a 256GB variant, so this will be something to look out for when choosing your next phone (like how 128GB models use UFS 3.1 now). That is a problem for next year, though – SK hynix says that it expects to win orders from smartphone makers this year and to start shipping in volume in the first three months of next year. It's not just smartphones, though, the company is also working on 321-layer designs for SSDs for consumers and data centers. Source
Yahoo
21-05-2025
- Business
- Yahoo
SK hynix Develops UFS 4.1 Solution Based on 321-High NAND
Optimized for on-device AI with best-in-class sequential reading performance, low power requirement Thickness reduced by 15% to fit into ultra-slim flagship smartphones Portfolio with world's highest 321-layer product to enhance SK hynix's leadership as full stack AI memory provider SEOUL, South Korea, May 21, 2025 /PRNewswire/ -- SK hynix Inc. (or "the company", announced today that it has developed UFS 4.1 solution product adopting the world's highest 321-layer 1Tb triple level cell 4D NAND flash for mobile applications. The development comes amid increasing requirements for high performance and low power of a NAND solution product to ensure a stable operation of on-device AI. The company expects the UFS 4.1 product, optimized for AI workload, to help enhance its memory leadership in the flagship smartphone markets. With an increase in demand for on-device AI leading to greater importance of the balance between computation capabilities and battery efficiency of a device, the mobile market is now requiring thinness and low power from a mobile device. The latest product comes with a 7% improvement in power efficiency, compared with the previous generation based on 238-high NAND and a slimmer 0.85mm thickness, down from 1mm before, to fit into a ultra-slim smartphone. The product also supports data transfer speed of 4300MB/s, the fastest sequential read* for a fourth-generation of UFS, while providing the best-in-class performance by also improving random read and write speed**, critical for multitasking, by 15% and 40%, respectively. Immediate provision of the required data for on-device AI and faster running speed and the responsivity of an application are expected to enhance user experience. * Sequential Read/Write: speed to read and write data of a file sequentially * Random Read/Write: speed to read and write data of dispersed files SK hynix plans to win customer qualification within the year and ship in volume from the first quarter of next year. The product will be provided in two capacity types – 512GB and 1TB. Ahn Hyun, President and Chief Development Officer, said that SK hynix plans to complete development of the 321-high 4D NAND-based SSD for both consumers and data centers within the year. "We are on track to expand our position as a full-stack AI memory provider in the NAND space by building a product portfolio with AI technological edge." About SK hynix Inc. SK hynix Inc., headquartered in Korea, is the world's top tier semiconductor supplier offering Dynamic Random Access Memory chips ("DRAM") and flash memory chips ("NAND flash") for a wide range of distinguished customers globally. The Company's shares are traded on the Korea Exchange, and the Global Depository shares are listed on the Luxembourg Stock Exchange. Further information about SK hynix is available at View original content to download multimedia: SOURCE SK hynix Inc.
Yahoo
16-05-2025
- Business
- Yahoo
InPsytech Joins Samsung SAFE™ IP Partner Program for Excellence in ONFI and UCIe IP Solutions
HSINCHU, May 16, 2025 /PRNewswire/ -- InPsytech, a leading provider of advanced silicon intellectual property (IP) solutions, also a member of Egis (6452 TWO) Group, is proud to announce that it has joined the Samsung SAFE™ IP Program as a partner, one of the highest honors granted to ecosystem partners within Samsung's Advanced Foundry Ecosystem (SAFE™). This recognizes InPsytech's exceptional contributions in delivering cutting-edge ONFI (Open NAND Flash Interface). InPsytech will expand IP portfolio in Samsung Foundry including UCIe IP solutions for multi-die applications and UCIe (Universal Chiplet Interconnect Express) IP solutions. The Samsung SAFE™ program is designed to foster close collaboration between Samsung Foundry and its partners, encouraging innovation and efficiency across the semiconductor design ecosystem. By honoring outstanding IP providers, Samsung highlights those who enable next-generation semiconductor performance and integration. InPsytech's ONFI IP enables high-speed, low-power NAND flash memory interfaces, widely adopted in storage and embedded applications which owns the latest version SCA (SEPARATE COMMAND ADDRESS) validation, DFE functionality support, Intelligent power management and Complete PHY+IO solution. Meanwhile, the UCIe IP solution provides high-bandwidth, low-latency chiplet-to-chiplet interconnect capabilities, facilitating seamless integration in advanced multi-die architectures and accelerating the adoption of chiplet-based system design. "We are truly honored to have been recognized as the Samsung SAFE™ IP Partner," Grace Leu, VP of Sales & Marketing of InPsytech said. "This recognition affirms our dedication to delivering industry-leading IP that meets the most demanding performance and reliability standards. We look forward to continued collaboration with Samsung to drive innovation across global semiconductor markets." Regarding this collaboration, Ben Hyo Gyuem Rhew, Vice President and Head of the Foundry IP Development Team at Samsung Electronics, stated, "This partnership highlights the strong synergy between both parties in terms of technology and market strategy. We are pleased to see this cooperation unfold and look forward to creating greater innovation together." The recognition further solidifies InPsytech's position as a trusted partner in the global semiconductor ecosystem and reinforces its commitment to advancing the design and scalability of future chip solutions. About InPsytech InPsytech is a Taiwan-based silicon IP company specializing in high-performance interface, memory, and interconnect IP solutions. The company is dedicated to empowering SoC designers through robust, scalable IP optimized for applications in AI, data centers, mobile, and edge computing. InPsytech partners with global foundries and design ecosystems to deliver IP solutions that accelerate time-to-market and enable next-generation semiconductor innovations. For more information, please visit View original content: SOURCE InPsytech
