@techreport{li-cats-kv-cache-distribution-00,
    number =    {draft-li-cats-kv-cache-distribution-00},
    type =      {Internet-Draft},
    institution =   {Internet Engineering Task Force},
    publisher = {Internet Engineering Task Force},
    note =      {Work in Progress},
    url =       {https://datatracker.ietf.org/doc/draft-li-cats-kv-cache-distribution/00/},
    author =    {Zhiqiang Li and Zongpeng Du and Junjie Wang and Wei Cheng and Guoying Zhang and Xun Sun and Chunhao Zhao},
    title =     {{KV Cache Distribution for Distributed LLM Inference: Use Case and Requirements}},
    pagetotal = 6,
    year =      2026,
    month =     jul,
    day =       4,
    abstract =  {In large language model (LLM) inference, the key-value (KV) cache holds the attention state computed from previously processed tokens. Reusing cached state across requests avoids repeated prefill computation and reduces time-to-first-token. In distributed inference deployments, the KV cache becomes a network-distributed resource: the effectiveness of steering a request to a service instance depends not only on computing and network metrics but also on whether reusable cached state is available at or near that instance. This document describes the KV cache distribution use case for Computing-Aware Traffic Steering (CATS), identifies the gaps relative to the existing CATS framework and metrics, and states requirements for cache-state metric exposure and for the distribution and synchronization of cached content across multiple cache tiers.},
}