2 papers
cs.AR2025
Sangam: Chiplet-Based DRAM-PIM Accelerator with CXL Integration for LLM Inferencing
Khyati Kiyawat, Zhenxing Fan, Yasas Seneviratne +5
Large Language Models (LLMs) are becoming increasingly data-intensive due to growing model sizes, and they are becoming memory-bound as the context length and, consequently, the ke…
cs.AR2025
Membrane: Accelerating Database Analytics with Bank-Level DRAM-PIM Filtering
Akhil Shekar, Kevin Gaffney, Martin Prammer +9
In-memory database query processing frequently involves substantial data transfers between the CPU and memory, leading to inefficiencies due to Von Neumann bottleneck. Processing-i…