이어진 연구실

Focus on next-generation memory architectures such as Processing-in-Memory (PIM) to overcome memory bottlenecks in large-scale AI and data-intensive applications. PIM enables computation directly within memory devices, reducing data movement overhead.

Design custom acceleration strategies for specialized applications like Fully Homomorphic Encryption (FHE) using NPUs or memory-centric architectures. Focus on optimizing FHE workloads for commercially available NPUs and PIM platforms.

Investigate system-level techniques to efficiently manage complex modern memory systems with heterogeneous components such as CXL-attached memory, NUMA-aware processors, and tiered memory hierarchies.