ARK: Avoiding Routing Collisions for KV Cache Transfer in Disaggregated LLM Inference
Proceedings of the ACM SIGCOMM 2026 Conference, pp. 2067–2073, 2026.
Abstract & citation for ARK: Avoiding Routing Collisions for KV Cache Transfer in Disaggregated LLM Inference
ARK coordinates network paths for KV-cache transfers in disaggregated LLM inference, reducing routing collisions without changes to network switches.
Hung-Chun Lin, Ting-Wei Hsu, Chung-En Ho, and Ahmed Saeed. "ARK: Avoiding Routing Collisions for KV Cache Transfer in Disaggregated LLM Inference." Proceedings of the ACM SIGCOMM 2026 Conference, pp. 2067–2073, 2026. DOI: 10.1145/3789240.3828750.
