A convergence of memory with some of the computing functions of the CPU is the only way to extract more performance from the hardware going forward, one leading tech CEO has claimed,
Seok-Hee Lee, CEO of SK Hynix, the world’s second-largest memory manufacturer behind Samsung, shared his predictions at the Institute of Electrical and Electronics Engineers’ International Reliability Physics Symposium (IRPS).
“There will be a convergence of memory and logic. The concept is to add a few computing functions of CPU to DRAM,” said Lee as reported by The Register.
Computing in memory
Lee also gave details about the evolution of memory from the currency generation of high-bandwidth memory (HBM).
The next evolution, Processing Near Memory (PNM) will bring both the CPU and the memory within a single module. To further improve the performance, the CPU and the memory will be placed within a single package called Processing in Memory (PIM).
For maximum performance though, we’ll have to look towards Computing in Memory (CIM) where the CPU and the memory are integrated within the same die.
Lee went as far as to suggest that this merger can only happen with increased collaboration between the companies in the semiconductor industry ecosystem.
According to The Register, Lee spent a lot of time talking about SK Hynix’s upcoming breakthroughs that will help the company deliver faster, higher-capacity memory.
“We are improving materials and design structures for technical evolution in each field of DRAM and NAND, and solving the reliability problems step by step. If the platform is innovated successfully based on this, it is possible to achieve the DRAM process below 10 nanometers (nm) and stack over 600 layers of NAND in the future,” Lee said.
This will be a significant bump up for SK Hynix that currently leads the industry with its 176-layer NAND flash.
Via: The Register