# Bridging LLM Serving and CXL-SSDs with Chunk-Aware KV Cache Management

arXiv: https://arxiv.org/abs/2609.26828 · HTML: https://mlsanity.com/paper/2609.26828/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 46 references that are in the map.
- First posted on arXiv 20 Sep 2026.

## Builds on (most cited first, 27 of 27)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints](https://mlsanity.com/paper/2305.13245/) (2023)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization](https://mlsanity.com/paper/2401.18079/) (2024)
- [Splitwise: Efficient generative LLM inference using phase splitting](https://mlsanity.com/paper/2311.18677/) (2023)
- [RepoCoder: Repository-Level Code Completion Through Iterative Retrieval and Generation](https://mlsanity.com/paper/2303.12570/) (2023)
- [InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management](https://mlsanity.com/paper/2406.19707/) (2024)
- [CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion](https://mlsanity.com/paper/2405.16444/) (2024)
- [CacheGen: KV Cache Compression and Streaming for Fast Large Language Model Serving](https://mlsanity.com/paper/2310.07240/) (2023)
- [Mooncake: A KVCache-centric Disaggregated Architecture for LLM Serving](https://mlsanity.com/paper/2407.00079/) (2024)
- [Cost-Efficient Large Language Model Serving for Multi-turn Conversations with CachedAttention](https://mlsanity.com/paper/2403.19708/) (2024)
- [How to Train Long-Context Language Models (Effectively)](https://mlsanity.com/paper/2410.02660/) (2024)
- [LMCache: An Efficient KV Cache Layer for Enterprise-Scale LLM Inference](https://mlsanity.com/paper/2510.09665/) (2025)
- [A Comprehensive Survey on Long Context Language Modeling](https://mlsanity.com/paper/2503.17407/) (2025)
- [RAGCache: Efficient Knowledge Caching for Retrieval-Augmented Generation](https://mlsanity.com/paper/2404.12457/) (2024)
- [Stateful Large Language Model Serving with Pensieve](https://mlsanity.com/paper/2312.05516/) (2023)
- [POD-Attention: Unlocking Full Prefill-Decode Overlap for Faster LLM Inference](https://mlsanity.com/paper/2410.18038/) (2024)
- [NEO: Saving GPU Memory Crisis with CPU Offloading for Online LLM Inference](https://mlsanity.com/paper/2411.01142/) (2024)
- [TraCT: Disaggregated LLM Serving with CXL Shared Memory KV Cache at Rack-Scale](https://mlsanity.com/paper/2512.18194/) (2025)
- [Beluga: A CXL-Based Memory Architecture for Scalable and Efficient LLM KVCache Management](https://mlsanity.com/paper/2511.20172/) (2025)
- [SkyByte: Architecting an Efficient Memory-Semantic CXL-based SSD with OS and Hardware Co-design](https://mlsanity.com/paper/2501.10682/) (2025)
- [ITME: Inference Tiered Memory Expansion with Disaggregated CXL-Hybrid Memories](https://mlsanity.com/paper/2606.12556/) (2026)
- [Performance Characterizations and Usage Guidelines of Samsung CXL Memory Module Hybrid Prototype](https://mlsanity.com/paper/2503.22017/) (2025)
- [CXL Topology-Aware and Expander-Driven Prefetching: Unlocking SSD Performance](https://mlsanity.com/paper/2505.18577/) (2025)
- [OpenCXD: An Open Real-Device-Guided Hybrid Evaluation Framework for CXL-SSDs](https://mlsanity.com/paper/2508.11477/) (2025)

## Abstract (arXiv)

NAND-backed storage offers the capacity needed to scale LLM prefix caching, but its block I/O path incurs CPU cache contention and host-DRAM staging in addition to NAND latency. Our characterization shows that these interface costs persist even with DRAM as the storage medium, motivating CXL-SSDs for byte-addressable access to NAND-backed capacity. Surprisingly, however, a stock CXL-SSD remains about 3× slower than local DRAM and no faster than an NVMe SSD, while generic prefetching provides little benefit. We present LM-CXD, a CXL-SSD specialized for LLM prefix caching. LM-CXD bridges the semantic gap between the serving engine, which knows which KV chunks will be consumed, and the device, which controls their placement and movement. It makes KV chunks device-visible I/O units, exposes NAND-to-DRAM progress to the serving engine, and uses device DRAM as a GPU-accessible buffer. LM-CXD further coordinates request scheduling with windowed prefetching and pipelines layerwise KV movement with GPU computation to hide NAND latency under limited device DRAM. Across five LLM models, LM-CXD reduces average TTFT over a stock CXL-SSD by up to 2.6× with compute asynchronous prefetching and 4.03× with layerwise prefetching, achieving TTFT within 1.5× of local DRAM on average.