Samsung Electronics said it has started mass production of its PM1763 enterprise solid-state drive (SSD), aiming to meet growing demand for high-speed storage as artificial intelligence workloads drive investment in next-generation data centre infrastructure.
The PM1763 is based on the PCIe 6.0 interface and is designed for enterprise servers used in AI training, inference and high-performance computing (HPC), the company said.
The launch comes as technology companies expand AI infrastructure to support increasingly large language models and data-intensive applications, boosting demand for faster storage capable of reducing data transfer bottlenecks between processors and accelerators.
Samsung said the drive combines its ninth-generation V-NAND flash memory with a 4-nanometre controller developed in-house to improve data throughput and energy efficiency.
The PM1763 will be offered in 4-terabyte (TB), 8TB and 16TB capacities. According to the company, the 16TB model delivers sequential read speeds of up to 28,400 megabytes per second (MB/s) and write speeds of up to 21,900MB/s, more than twice the performance of its previous-generation PM1753 SSD.
Samsung said the performance enables a 40-gigabyte large language model to be transferred in about 1.4 seconds, reducing latency in AI computing 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,” Jangseok Choi, Vice President and Head of Memory Product Planning at Samsung Electronics, said in a statement.
The company said the SSD has been optimised for liquid-cooled servers through direct-to-chip cooling technology, enabling sustained performance during intensive computing workloads.
Samsung also said the PM1763 delivers more than 1.8 times the power efficiency of its predecessor, a feature aimed at reducing energy consumption and operating costs for hyperscale data centres.
The drive incorporates support for post-quantum cryptography (PQC) algorithms, which are designed to protect data against future quantum computing threats, as well as the TEE Device Interface Security Protocol (TDISP) for securing data movement in virtualised computing environments.
Enterprise SSD manufacturers have been introducing faster storage technologies as cloud service providers and AI infrastructure operators scale investments in advanced computing systems. High-performance storage has become a key component of AI clusters, where rapid movement of large datasets is critical to model training and inference.
The PM1763 expands Samsung’s enterprise storage portfolio as competition intensifies among memory manufacturers seeking to supply hardware for AI-driven data centres.
