Beyond Compression Ratio: High-Throughput KV Compression for Disaggregated LLM Serving
cs.DC
Submitted: 2026-11-03
Updated: 2026-11-04
Comments: Accepted by The 22nd Annual IFIP International Conference on Network and Parallel Computing (NPC 2026)
License: http://creativecommons.org/licenses/by-sa/4.0/
The gist: KV cache compression reduces the communication volume of disaggregated large language model serving, but its practical benefit depends not only on compression ratio and model quality…
Terminology
Abstract
KV cache compression reduces the communication volume of disaggregated large language model serving, but its practical benefit depends not only on compression ratio and model quality, but also on how efficiently the compression pipeline executes. This requirement becomes increasingly important as modern serving systems adopt high-bandwidth interconnects, where inefficient compression can offset a substantial portion of the saved communication time. This paper presents FastKVComp, a high-throughput KV cache compression data path built on top of KVServe. FastKVComp improves compression execution in two complementary ways. First, it fuses the distribution transformation and low-bit quantization stages into a KVaware TileLang kernel, eliminating intermediate tensor materialization and redundant memory accesses. Second, it replaces the original lossless compression backend with an open, extensible, and searchable backend based on LC, enabling high-performance compression operator pipelines. The fused kernel achieves up to 5.9x speedup, while the new lossless backend increases compression throughput from below 90 GB/s to over 500 GB/s. Across the evaluated workloads, FastKVComp reduces job completion time by up to 2.04x. End-to-end evaluation shows that these optimizations allow KV compression to retain meaningful communication benefits under modern high-bandwidth network.
Sources
- The Llama 3 Herd of Models
- SplitZip: Ultra Fast Lossless KV Compression for Disaggregated LLM Serving
- GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM
- SpinQuant: LLM quantization with learned rotations
- KVServe: Service-Aware KV Cache Compression for Communication-Efficient Disaggregated LLM Serving
- TileLang: A Composable Tiled Programming Model for AI Systems
- Qwen2.5 Technical Report
- No Token Left Behind: Reliable KV Cache Compression via Importance-Aware Mixed Precision Quantization
Related papers
- Programmable Probabilistic Computer with 1,000,000 p-bits
- BatchGen: An Architecture for Scalable and Efficient Batch Inference
- Seamless Reconfiguration for DAG-BFT (Technical Report)
- ELDR: Expert-Locality-Aware Decode Routing for PD-Disaggregated MoE Serving
- iScheduler: Reinforcement Learning-Driven Continual Optimization for Large-Scale Resource Investment Problems
- TIRx: A Unified Tile Primitive ML Compiler Across Heterogeneous Backends