Requests of a Feather Must Flock Together: Batch Size vs. Prefix Homogeneity in LLM Inference
Auto-regressive token generation in large language models is memory-bound because it requires "attending to" key and value tensors (KV cache) of all previous tokens. Prior work aims to improve the efficiency of this decode process by batching multiple requests together, and maximizing batch size subject to GPU memory constraints. The key observation of our work is that with prefix-sharing workloads, smaller, prefix-homogeneous batches - where all requests share a common prefix - can achieve higher decode throughput than larger, heterogeneous batches, due to better spatial and temporal locality during KV cache accesses. However, prefix-aware schedulers in state-of-the-art inference engines...
| Authors | Saksham Rathi, Preeti, Mythili Vutukuru |
|---|---|
| Primary category | cs.LG |
| Published | 7 May 2026 |
| Last updated at source | 10 October 2026 |
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.