InKubeator: Pre-warming In-Memory KV Caches from Disk for Elastic LLM Serving

Abhishek Muthukumar, Erik Gough, Deepak Nadig. InKubeator: Pre-warming In-Memory KV Caches from Disk for Elastic LLM Serving. In IEEE International Conference on Communications, ICC 2026, Glasgow, United Kingdom, May 24-28, 2026. pages 1-6, IEEE, 2026. [doi]

Abstract

Abstract is missing.