Samsung develops industry’s first DDR5 memory built using HKMG process

Samsung, the world’s biggest memory chip maker, has announced that it has developed the industry’s first 512GB DDR5 DRAM module using the HKMG (High-K Metal Gate) process technology. The new memory module delivers up to double the performance of DDR4 DRAM chips at 7,200Mbps.

The HKMG process is typically used in manufacturing logic chips (processors), but Samsung is the first brand to use the process to manufacture DRAM chips. The company’s newest DDR5 memory module is suitable for some of the most compute-hungry devices such as supercomputers and servers that work on AI (Artificial Intelligence), ML (Machine Learning), data analytics, and medical research. The new DDR5 modules also use through-silicon via (TSV) technology to stack up to eight layers of 16Gb DRAM chips to offer a total capacity of 512GB.

Due to the continued scaling down of DRAM structures, the insulation layer gets thinned and leads to high current leakage. With Samsung’s new process, the insulating material gets replaced with HKMG, leading to lower current leakage and higher performance. The HKMG-based DDR5 DRAM modules consume 13% lower power, making them suitable for data centers where power efficiency matters.

The South Korean firm had earlier used the HKMG process to manufacture GDDR6 memory in 2018 for GPUs. Samsung Semiconductor is currently sampling its new DDR5 memory modules in different variations to customers for verification. Over the next few weeks, the modules will receive certification for final usage.

Young-Soo Sohn, Vice President of the DRAM Memory Planning/Enabling Group at Samsung Electronics, said, “Samsung is the only semiconductor company with logic and memory capabilities and the expertise to incorporate HKMG cutting-edge logic technology into memory product development. By bringing this type of process innovation to DRAM manufacturing, we are able to offer our customers high-performance, yet energy-efficient memory solutions to power the computers needed for medical research, financial markets, autonomous driving, smart cities, and beyond.

Join the Discussion