[REVISION HISTORY]

Kioxia enterprise SSD technology developments

Updated 3 times since CLSTR started tracking revisions of this situation.

What changed

2026-10-20 01:53 UTC → 2026-10-22 01:23 UTC · added ~~removed~~

Kioxia has introduced new enterprise SSD developments aimed at high-performance and high-capacity data center workloads. At the Open Compute Project Global Summit, the company unveiled its GP series, which utilizes third-generation XL-Flash technology to meet Nvidia’s Storage-Next initiative. Kioxia has established a long-term roadmap for these technologies, targeting an SSD with 100 million IOPS by 2028 using PCIe 7.0. In addition to the GP series, Kioxia announced the LD4 series, which employs eighth-generation BiCS8 QLC NAND flash and the E1.L form factor. These PCIe Gen5 NVMe SSDs are designed for read-intensive hyperscale workloads in high-density 1U scale-out servers. While the architecture is validated to support capacities up to 122.88TB, the company is currently shipping 15.36TB and 30.72TB samples to select customers for functional testing. The LD4 series supports PCIe 5.0 and the Open Compute Project (OCP) datacenter NVMe SSD specification v2.6. Expanding its enterprise offerings, Kioxia has announced new models in the CM9 SSD series designed for Direct Liquid Cooling (DLC). These E3.S form factor ~~drives~~ drives, utilizing 8th generation BiCS FLASH TLC memory, are engineered to support cold plate cooling in server designs while maintaining compatibility with traditional air-cooled environments. The CM9-DLC models aim to address the thermal demands and power consumption of high-density AI workloads, such as SSD context caching. These PCIe 5.0 and NVMe 2.0 compliant drives offer capacities up to 30.72TB for read-intensive use cases and 25.6TB for mixed-use scenarios. To ensure the quality of its next-generation PCIe 5.0 and PCIe 6.0 compatible SSDs, Kioxia is utilizing Anritsu’s MP1900A Signal Quality Analyzer. This collaboration focuses on high-speed interface quality assessments, including compliance testing and link training evaluation, to support the development of high-performance storage systems for AI servers and cloud services.

Versions

Only revisions since CLSTR began indexing content versions appear here. Select a version to see what changed compared to the one before it.