ITME uses CXL-hybrid memories for byte-addressable remote memory expansion in LLM inference, achieving up to 35.7% throughput improvement over conventional CPU-offloading.
Goumas, Zeshan Chishti, and Nandita Vijaykumar
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ITME: Inference Tiered Memory Expansion with Disaggregated CXL-Hybrid Memories
ITME uses CXL-hybrid memories for byte-addressable remote memory expansion in LLM inference, achieving up to 35.7% throughput improvement over conventional CPU-offloading.