Samsung Mass-Produces Enterprise SSDs for NVIDIA Vera Rubin
N.R. Finch
Samsung has begun mass-producing the PM1763 enterprise SSD for Nvidia's next-gen Vera Rubin AI platform — doubling read/write speeds and adding liquid cooling to lock in a key slot in the AI data-center supply chain.
Why does this drive matter?
The PM1763 is Samsung's latest data-center SSD, purpose-built for Nvidia's upcoming Vera Rubin platform.
This means → Samsung is not selling a generic drive; it has pre-secured the storage slot in Nvidia's next-gen AI servers.
The drive debuted at Nvidia's GTC conference alongside HBM4 high-bandwidth memory and low-power SOCAMM2 modules — three products forming a complete AI data-center package.
How much faster is it, and what drives the gain?
The PM1763 uses the latest V-NAND flash — vertically stacked storage chips — and a newly developed 4 nm controller, delivering read/write speeds more than double the prior generation.
In plain terms = a more advanced chip plus a new-generation "brain" managing the data adds up to twice the speed.
The drive also features liquid cooling, designed to sustain peak throughput during AI training and inference and cut data latency for processors and AI accelerators.
Where does Samsung stand in this market?
TrendForce data shows Samsung held 35% market share in enterprise SSDs in Q1 this year, ranking first.
Behind it: SK Hynix, Micron, Kioxia Holdings, and Kioxia partner SanDisk.
This means → Samsung is already the leader; whether this production ramp widens the gap is the key thing to watch next.
Content is for reference only, not financial advice.