# DUAL-BLADE: Dual-Path NVMe-Direct KV-Cache Offloading for Edge LLM Inference

arXiv: https://arxiv.org/abs/2604.26557 · HTML: https://mlsanity.com/paper/2604.26557/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in KV cache optimization, one of 12 directions in Efficient LLM inference and serving.
- Builds on 24 of its 45 references that are in the map.
- First posted on arXiv 29 Apr 2026.

## Placed in

- Efficient LLM inference and serving / [KV cache optimization](https://mlsanity.com/inference/kv-cache-optimization/)

## Builds on (most cited first, 24 of 24)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [Gemma: Open Models Based on Gemini Research and Technology](https://mlsanity.com/paper/2403.08295/) (2024)
- [FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU](https://mlsanity.com/paper/2303.06865/) (2023)
- [DeepSpeed Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale](https://mlsanity.com/paper/2207.00032/) (2022)
- [InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management](https://mlsanity.com/paper/2406.19707/) (2024)
- [MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases](https://mlsanity.com/paper/2402.14905/) (2024)
- [PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU](https://mlsanity.com/paper/2312.12456/) (2023)
- [LLM in a flash: Efficient Large Language Model Inference with Limited Memory](https://mlsanity.com/paper/2312.11514/) (2023)
- [Can Foundation Models Wrangle Your Data?](https://mlsanity.com/paper/2205.09911/) (2022)
- [Cost-Efficient Large Language Model Serving for Multi-turn Conversations with CachedAttention](https://mlsanity.com/paper/2403.19708/) (2024)
- [LMCache: An Efficient KV Cache Layer for Enterprise-Scale LLM Inference](https://mlsanity.com/paper/2510.09665/) (2025)
- [A Review on Edge Large Language Models: Design, Execution, and Applications](https://mlsanity.com/paper/2410.11845/) (2024)
- [LongReward: Improving Long-context Large Language Models with AI Feedback](https://mlsanity.com/paper/2410.21252/) (2024)
- [LogParser-LLM: Advancing Efficient Log Parsing with Large Language Models](https://mlsanity.com/paper/2408.13727/) (2024)
- [InstInfer: In-Storage Attention Offloading for Cost-Effective Long-Context LLM Inference](https://mlsanity.com/paper/2409.04992/) (2024)
- [A Cost-Benefit Analysis of On-Premise Large Language Model Deployment: Breaking Even with Commercial LLM Services](https://mlsanity.com/paper/2509.18101/) (2025)
- [GPU-Initiated On-Demand High-Throughput Storage Access in the BaM System Architecture](https://mlsanity.com/paper/2203.04910/) (2022)
- [KVSwap: Disk-aware KV Cache Offloading for Long-Context On-device Inference](https://mlsanity.com/paper/2511.11907/) (2025)

## Built on by (most cited first, 2 of 2)

- [When Cloud Agents Meet Device Agents: Lessons from Hybrid Multi-Agent Systems](https://mlsanity.com/paper/2605.30102/) (2026)
- [Building py-kvcache: A Performance Characterization of External KV Caching for vLLM with NVMe SSDs](https://mlsanity.com/paper/2609.11744/) (2026)

## Abstract (arXiv)

The increasing deployment of Large Language Model (LLM) inference on edge AI systems demands efficient execution under tight memory budgets. A key challenge arises from Key-Value (KV) caches, which often exceed available device memory. Although NVMe-based offloading offers scalable capacity, existing file-based designs rely heavily on the kernel page cache, leading to cache thrashing, unpredictable latency, and high software overhead under memory pressure. We present DUAL-BLADE, a dual-path KV residency framework that dynamically assigns KV tensors to either a page-cache path or an NVMe-direct path based on runtime memory availability. The NVMe-direct path bypasses the filesystem by mapping KV tensors to contiguous logical block address (LBA) regions, enabling low-overhead direct storage access. DUAL-BLADE further incorporates adaptive pipeline parallelism to overlap storage I/O with GPU DMA, improving inference throughput. Our evaluation shows that DUAL-BLADE substantially mitigates I/O bottlenecks, reducing prefill and decode latency by up to 33.1% and 42.4%, respectively, while improving SSD utilization by 2.2x across diverse memory budgets.