Mooncake
paper datasetA KVCache-centric disaggregated architecture for LLM serving. Won Best Paper at FAST 2025.
Outputs 2
Mooncake: KVCache-centric Disaggregated Architecture
paperarXiv: 2407.00079
Venue: FAST 2025
arXiv: 2407.00079
Venue: FAST 2025