Kioxia and SanDisk have unveiled a next-generation QLC 3D flash memory that improves storage density, bandwidth, and power efficiency.
The development targets global data centers, cloud providers, and AI developers struggling to handle rising data volumes as AI systems scale.
What happened
Kioxia and SanDisk have introduced a next-generation QLC 3D flash memory technology that increases storage density, bandwidth, and power efficiency.
Why it matters
The advancement is designed to help global data centers, cloud providers, and AI developers manage surging data demand, addressing a critical infrastructure need as AI workloads expand.
What to watch
The specific performance metrics, availability timeline, and deployment roadmap for this technology in production environments remain to be detailed.
Kioxia and SanDisk have jointly introduced a next-generation QLC 3D flash memory technology designed to address mounting storage pressures in AI-driven data centers. QLC (quad-level cell) flash stores four bits per cell, enabling higher density than earlier generations. The new iteration boosts three key parameters: storage density—the amount of data that can fit in a given physical space—bandwidth, the speed at which data moves in and out of the memory; and power efficiency, reducing the electrical overhead of storage operations. These improvements are aimed directly at global data centers, cloud providers, and AI developers, all of whom face surging data demand as AI models grow larger and more complex. The technology positions flash memory as a critical enabler for managing the data volumes that AI infrastructure now routinely processes, bridging the gap between compute performance and storage capacity that has become a bottleneck in large-scale AI deployments.
The announcement underscores a fundamental challenge in AI infrastructure: as AI models and workloads grow in scale and complexity, the storage systems that feed them must keep pace. Data centers and cloud providers increasingly rely on high-density, power-efficient flash memory to support both training and inference workloads without inflating operational costs. By focusing on QLC 3D flash—a technology that stores more data per cell than earlier generations—Kioxia and SanDisk are targeting one of the most cost-sensitive layers of the AI compute stack. The emphasis on bandwidth and power efficiency signals recognition that raw capacity alone is insufficient; modern AI infrastructure demands speed and thermal efficiency to maintain performance and sustainability at scale.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
Ask AI anything about this article. Q&As are published on this page for other readers too.
Supermicro reported a sharp recovery in gross margins and issued fiscal 2027 revenue guidance that exceeded Wa…
IBM and Together AI announced a collaboration to deploy a dedicated Nvidia-powered inference cluster on IBM Cl…
Siliconware Precision Industries (SPIL), a subsidiary of ASE Technology Holding, held a groundbreaking ceremon…
CoreWeave reported second-quarter fiscal 2026 revenue of US$2.6 billion, up 112% year-over-year and 24% sequen…
CoreWeave, an AI cloud provider, more than doubled its second-quarter revenue while managing a $104 billion ba…
Palantir Technologies reported second-quarter revenue up 93% year over year to $1.94 billion on Aug
The AI news that matters, in one minute each morning.
Sign up free