Samsung Electronics has announced mass production of the PM1763, the industry's first PCIe 6.0-based enterprise solid state drive optimized for next-generation AI and high-performance computing (HPC) server environments.

The announcement marks a critical milestone in the AI infrastructure arms race, as the exploding volume of data required for AI training and inference is driving demand for storage solutions that can keep pace with increasingly powerful processors and accelerators.

Blazing Speed for AI Workloads

The PM1763 delivers sequential read speeds of up to 28,400 MB/s and write speeds of up to 21,900 MB/s — more than double the performance of its predecessor, the PM1753. At this speed, a 40GB large language model (LLM) can be transferred in approximately 1.4 seconds, dramatically reducing data latency between processors and accelerators.

The drive is available in 4TB, 8TB, and 16TB capacities, with Samsung confirming it has completed validation for next-generation AI platforms — notably NVIDIA's Vera Rubin architecture, which is designed to offload AI context into massive pools of SSDs with each rack packing over 1,100 TB of NAND storage.

Built on Advanced Technology

Samsung's PM1763 combines the company's 9th-generation V-NAND with a newly developed 4-nanometer (nm) controller, delivering significant improvements in both performance and power efficiency. Power efficiency has improved by more than 1.8 times compared to the previous generation, helping reduce datacenter operating costs.

For demanding AI server environments, the PM1763 is optimized for liquid-cooled deployments through direct-to-chip (D2C) cooling technology, enabling sustained peak performance under intensive workloads and extended operating conditions.

Security for the Post-Quantum Era

In a forward-looking move, Samsung has equipped the PM1763 with post-quantum cryptography (PQC) algorithms designed to protect against future quantum computing threats. The drive also supports TEE Device Interface Security Protocol (TDISP), which helps secure data pathways in virtualized environments.

"Built on industry-leading performance, PM1763 has successfully completed validation for next-generation AI platforms and is well positioned to support evolving AI infrastructure requirements," said Jangseok Choi, Vice President and Head of Memory Product Planning at Samsung Electronics. "As AI models continue to grow in size and complexity, PM1763 will serve as a key solution that enables customers to efficiently scale memory capacity and optimize AI operations."

The PM1763 positions Samsung at the center of the AI hardware ecosystem, as companies race to build the infrastructure needed to support the next generation of massive AI models.

Sources