"Performing 'Storage and Computation' Together... SK Hynix Develops Next-Generation Memory Semiconductor PIM"

Maximizing Data Processing Speed... Reducing Energy Consumption by 80%
Prototype Applying PIM to Be Unveiled at World's Leading Conference

[Asia Economy Reporter Lee Hye-young] SK Hynix announced on the 16th that it has developed next-generation memory semiconductors with processing capabilities (PIM·Processing-In-Memory). SK Hynix plans to accelerate the development of next-generation semiconductors equipped with both storage and processing functions, building a converged solution ecosystem that surpasses the limitations of the memory domain.

PIM is a next-generation technology that adds processing functions to memory semiconductors, addressing data transfer bottlenecks in artificial intelligence (AI) and big data processing fields.

Until now, memory semiconductors have been responsible for data storage, while processing functions, similar to those of the human brain, have been handled by non-memory semiconductors such as central processing units (CPU) or graphics processing units (GPU). However, recently, global semiconductor companies have been releasing converged products that break the boundaries between memory and system semiconductors, intensifying technological competition.

SK Hynix has also been steadily researching next-generation smart memory products to expand the memory semiconductor domain, and this time has unveiled its first result.

SK Hynix plans to disclose its PIM development achievements at the '2022 International Solid-State Circuits Conference (ISSCC),' the world's most prestigious semiconductor conference, held in San Francisco, USA, at the end of this month. The industry expects that as this technology evolves, 'memory-centric computing,' where memory semiconductors play a central role in information and communication technology (ICT) devices such as smartphones, will become possible.

Along with this, SK Hynix has developed a sample of its first product applying PIM, called 'GDDR6-AiM.' This product adds processing functions to GDDR6 memory, which processes data at a speed of 16 gigabits per second (Gbps). When this product is installed together with CPU and GPU instead of conventional DRAM, the speed of specific computations can increase up to 16 times. In the future, GDDR6-AiM is expected to be used in machine learning, high-performance computing, big data processing, and storage.

Notably, this product operates at 1.25V, which is lower than the existing operating voltage of GDDR6 at 1.35V. Additionally, the PIM performing its own processing reduces data transfers to the CPU and GPU, thereby lowering power consumption. As a result, energy consumption can be reduced by approximately 80% compared to existing products.

SK Hynix expects this to reduce carbon emissions from devices using the product, achieving ESG (Environmental, Social, and Governance) management effects.

SK Hynix plans to further accelerate technology development to expand the memory domain, including intelligent semiconductors. To this end, it recently announced plans to collaborate with Sapeon, an AI semiconductor company spun off from SK Telecom, to showcase technology combining GDDR6-AiM and AI semiconductors.

Ryu Su-jeong, CEO of Sapeon, said, "The use of artificial neural network data has rapidly increased recently, requiring computing technologies optimized for such processing characteristics. By integrating the technologies of both companies, we will maximize efficiency in terms of data processing, cost, and energy usage."

Ahn Hyun, Vice President of Solution Development at SK Hynix, said, "SK Hynix will build a new memory solution ecosystem using PIM-based GDDR6-AiM with built-in processing capabilities. We will continue to evolve our business model and technology development direction."

This content was produced with the assistance of AI translation services.