Enterprise SSD accelerates AI server data transfer
Samsung’s PM1763 PCIe 6.0-based enterprise SSD features 9th-generation V-NAND flash memory and a new 4-nm controller. Optimized for AI and HPC servers, the drive is available in 4-TB, 8-TB, and 16-TB capacities. The 16-TB model delivers sequential read and write speeds of up to 28,400 MB/s and 21,900 MB/s, respectively—up to twice the performance of its predecessor, the PM1753.
According to the company, the PM1763 can transfer a 40-GB LLM in approximately 1.4 seconds, helping minimize data latency between processors and accelerators while improving overall AI processing efficiency. The SSD is optimized for liquid-cooled server environments through direct-to-chip cooling. This design enables sustained peak performance while improving power efficiency by up to 1.8 times compared to the previous generation.
To address security requirements for AI and virtualized infrastructure, the PM1763 supports post-quantum cryptography (PQC), the Security Protocol and Data Model (SPDM) 1.4, and Commercial National Security Algorithm (CNSA) 2.0. It also provides link encryption based on the TEE Device Interface Security Protocol (TDISP) to reinforce data protection across storage interfaces.
Samsung has now begun mass production of the PM1763 SSD.
The post Enterprise SSD accelerates AI server data transfer appeared first on EDN.
