FlashOffloading: Full KV Cache in Host DRAM for Long-Context Serving

This episode examines FlashOffloading, a system that keeps the entire KV cache in host DRAM during both prefill and decode for long-context serving on Ascend 910C chips. It starts with the memory problem: GLM-5.2 needs about 95 KB of KV per token, so a single million-token request needs roughly 97.5 GB. Each decode die has only about 28 GB for KV, so HBM limits concurrency before compute does. The discussion explains why sparse attention with a native top-2048 token indexer makes offloading practical. Each step touches only a small subset of tokens, and attention over them stays exact. The episode then covers the three-part design: rolling layer buffers for prefill, a reverse index with a fused kernel for decode cache management, and a kernel that overlaps miss fetches with attention on hits. It also tests the headline claims of 1.38x prefill and 1.91x decode throughput under a 30 ms per-token budget by asking what the baselines were allowed to do. Listeners get a clear explanation of TTFT, TPOT, ITL, speculative decoding and prefill-decode disaggregation, and of how this work differs from earlier offloading systems such as FlexGen, Quest and Mooncake. Sources: 1. FlashOffloading: Low-Latency KV Cache Offloading for Long-Context LLM Serving — Xuan Wang, Zhuanhao Wu, Tingzhu Fang, Xushi Li, Sicheng Zhang, Zhenyu Yang, Feng Liu, Zhaohui Fan, Qiang Xie, Huayi Zou, Yuncong Liu, Zeyu Song, Ke Zhu, Yanjia Zeng, Jingchun Gao, Yan Shi, Kui Luo, Gezi Li, Liangjun Feng, Cheng Li, 2026 http://arxiv.org/abs/2611.07542 2. FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU — Ying Sheng, Lianmin Zheng, Binhang Yuan, et al., 2023 https://scholar.google.com/scholar?q=FlexGen%3A+High-Throughput+Generative+Inference+of+Large+Language+Models+with+a+Single+GPU 3. FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision — Jay Shah, Ganesh Bikshandi, Ying Zhang, Vijay Thakkar, Pradeep Ramani, Tri Dao, 2024 https://scholar.google.com/scholar?q=FlashAttention-3%3A+Fast+and+Accurate+Attention+with+Asynchrony+and+Low-precision 4. DeepSeek-V3.2-Exp: Boosting Long-Context Efficiency with DeepSeek Sparse Attention — DeepSeek-AI, 2025 https://scholar.google.com/scholar?q=DeepSeek-V3.2-Exp%3A+Boosting+Long-Context+Efficiency+with+DeepSeek+Sparse+Attention 5. Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference — Jiaming Tang, Yilong Zhao, Kan Zhu, Guangxuan Xiao, Baris Kasikci, Song Han, 2024 https://scholar.google.com/scholar?q=Quest%3A+Query-Aware+Sparsity+for+Efficient+Long-Context+LLM+Inference 6. Mooncake: A KVCache-centric Disaggregated Architecture for LLM Serving — Ruoyu Qin, Zheming Li, Weiran He, et al. (Moonshot AI), 2024 https://scholar.google.com/scholar?q=Mooncake%3A+A+KVCache-centric+Disaggregated+Architecture+for+LLM+Serving 7. ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning — Samyam Rajbhandari, Olatunji Ruwase, Jeff Rasley, Shaden Smith, Yuxiong He, 2021 https://scholar.google.com/scholar?q=ZeRO-Infinity%3A+Breaking+the+GPU+Memory+Wall+for+Extreme+Scale+Deep+Learning 8. Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention — Jingyang Yuan et al. (DeepSeek-AI), 2025 https://scholar.google.com/scholar?q=Native+Sparse+Attention%3A+Hardware-Aligned+and+Natively+Trainable+Sparse+Attention 9. InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management — Wonbeom Lee, Jungi Lee, Junghwan Seo, Jaewoong Sim, 2024 https://scholar.google.com/scholar?q=InfiniGen%3A+Efficient+Generative+Inference+of+Large+Language+Models+with+Dynamic+KV+Cache+Management 10. LMCache: An Efficient KV Cache Layer for Enterprise-Scale LLM Inference — Yuhan Liu et al., 2025 https://scholar.google.com/scholar?q=LMCache%3A+An+Efficient+KV+Cache+Layer+for+Enterprise-Scale+LLM+Inference 11. Mooncake: Trading More Storage for Less Computation — A KVCache-centric Architecture for Serving LLM Chatbot — Ruoyu Qin et al., 2025 https://scholar.google.com/scholar?q=Mooncake%3A+Trading+More+Storage+for+Less+Computation+%E2%80%94+A+KVCache-centric+Architecture+for+Serving+LLM+Chatbot 12. ECHO: Efficient KV Cache Offloading with Lossless Prefetching for Serving Native Sparse Attention LLMs — Guangda Liu et al., 2026 https://scholar.google.com/scholar?q=ECHO%3A+Efficient+KV+Cache+Offloading+with+Lossless+Prefetching+for+Serving+Native+Sparse+Attention+LLMs 13. HiSparse: Scaling Sparse-Attention Decoding with Hierarchical KV Cache Management — Xie, Huang, Huang, Xu, Ma, Kozyrakis, 2026 https://scholar.google.com/scholar?q=HiSparse%3A+Scaling+Sparse-Attention+Decoding+with+Hierarchical+KV+Cache+Management 14. Understanding Bottlenecks for Efficiently Serving LLM Inference with KV Offloading — Meng, Lee, Wang, 2025 https://scholar.google.com/scholar?q=Understanding+Bottlenecks+for+Efficiently+Serving+LLM+Inference+with+KV+Offloading Interactive Visualization: FlashOffloading: Full KV Cache in Host DRAM for Long-Context Serving