TokenDance: Scaling Multi-Agent LLM Serving via Collective KV Cache Sharing
This addresses scalability bottlenecks for multi-agent AI applications by reducing communication and storage overheads, representing a novel system-level optimization rather than an incremental improvement.
The paper tackles the problem of KV cache redundancy in multi-agent LLM serving by introducing TokenDance, a system that enables collective KV cache sharing, resulting in up to 2.7x more concurrent agents, 17.5x storage reduction, and 1.9x prefill speedup.
Multi-agent LLM applications organize execution in synchronized rounds where a central scheduler gathers outputs from all agents and redistributes the combined context. This All-Gather communication pattern creates massive KV Cache redundancy, because every agent's prompt contains the same shared output blocks, yet existing reuse methods fail to exploit it efficiently. We present TokenDance, a system that scales the number of concurrent agents by exploiting the All-Gather pattern for collective KV Cache sharing. TokenDance's KV Collector performs KV Cache reuse over the full round in one collective step, so the cost of reusing a shared block is paid once regardless of agent count. Its Diff-Aware Storage encodes sibling caches as block-sparse diffs against a single master copy, achieving 11-17x compression on representative workloads. Evaluation on GenerativeAgents and AgentSociety shows that TokenDance supports up to 2.7x more concurrent agents than vLLM with prefix caching under SLO requirement, reduces per-agent KV Cache storage by up to 17.5x, and achieves up to 1.9x prefill speedup over per-request position-independent caching.