MMA routes host-GPU transfers over multiple available paths to deliver 4.62x higher peak bandwidth and lower latencies in LLM serving without hardware or driver changes.
Prompt cache: Modular attention reuse for low-latency inference.Pro- ceedings of Machine Learning and Systems, 6:325–338
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
MultiPath Memory Access: Breaking Host-GPU Bandwidth Bottlenecks in LLM Services
MMA routes host-GPU transfers over multiple available paths to deliver 4.62x higher peak bandwidth and lower latencies in LLM serving without hardware or driver changes.