SK hynix has created a new memory chip that includes processing in memory (PIM), a computer chip with computational abilities. In addition to computational functions to semiconductor memory, PIM provides a solution for artificial intelligence (AI) and big data congestion.
Ryu Soo-Jung, CEO of SAPEON Inc., said, “The use of artificial neural network data has increased rapidly recently, requiring computing technology optimized for computational characteristics. We aim to maximize efficiency in data calculation, costs, and energy use by combining technologies from the two companies.”
Ahn Hyun, head of Solution Development who spearheaded the development of the latest technology and product, said that “SK hynix will build a new memory solution ecosystem using GDDR6-AiM, which has its own computing function.” He added that “the company will continue to evolve its business model and the direction for technology development.”
The general consensus is that memory chips store data, but the CPU or GPU, like the human brain, processes that data. The latest technology from SK Hynix has led to the development of a breakthrough solution.
As part of the world’s most prestigious semiconductor conference, ISSCC 2022, which is taking place at the end of this month in San Francisco, SK Hynix plans to present its PIM technology. The company anticipates that efforts to innovate this technology will continue, bringing memory-centric computing, in which semiconductor memory plays a central role, one step closer to reality in devices such as smartphones.
SK hynix has developed a sample of GDDR6-AiM as the first product to use PIM technology (Accelerator in memory). The GDDR6-AiM adds computational functions to 16Gbps GDDR6 memory chips. When GDDR6-AiM is used in conjunction with a CPU or GPU instead of traditional DRAM, a certain computation speed can be increased by 16 times. GDDR6-AiM is expected to be used in machine learning, high-performance computing, and big data computation and storage.
The operating voltage of GDDR6-AiM is 1.25V, which is lower than the existing product’s operating voltage of 1.35V. Furthermore, the PIM reduces data movement to the CPU and GPU, resulting in an 80 percent reduction in power consumption. SK hynix has the capability to fulfill its commitment to ESG management by lowering carbon emissions from devices that use this product.
In collaboration with SAPEON Inc., an AI chip company that was recently spun off from SK Telecom, SK hynix also plans to introduce a technology that combines GDDR6-AiM with AI chips.