# Strata: Hierarchical Context Caching for Long Context Language Model Serving

arXiv: https://arxiv.org/abs/2508.18572 · HTML: https://mlsanity.com/paper/2508.18572/

## Key facts

- Cited by 26 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 15 of its 40 references that are in the map.
- First posted on arXiv 26 Aug 2025.

## Placed in

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

## Builds on (most cited first, 15 of 15)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [SGLang: Efficient Execution of Structured Language Model Programs](https://mlsanity.com/paper/2312.07104/) (2023)
- [The NarrativeQA Reading Comprehension Challenge](https://mlsanity.com/paper/1712.07040/) (2017)
- [Prompt Cache: Modular Attention Reuse for Low-Latency Inference](https://mlsanity.com/paper/2311.04934/) (2023)
- [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)
- [Cost-Efficient Large Language Model Serving for Multi-turn Conversations with CachedAttention](https://mlsanity.com/paper/2403.19708/) (2024)
- [NanoFlow: Towards Optimal Large Language Model Serving Throughput](https://mlsanity.com/paper/2408.12757/) (2024)
- [MemServe: Context Caching for Disaggregated LLM Serving with Elastic Memory Pool](https://mlsanity.com/paper/2406.17565/) (2024)
- [Stateful Large Language Model Serving with Pensieve](https://mlsanity.com/paper/2312.05516/) (2023)
- [Peer Review as A Multi-Turn and Long-Context Dialogue with Role-Based Interactions](https://mlsanity.com/paper/2406.05688/) (2024)
- [Compute Or Load KV Cache? Why Not Both?](https://mlsanity.com/paper/2410.03065/) (2024)
- [GPU-Initiated On-Demand High-Throughput Storage Access in the BaM System Architecture](https://mlsanity.com/paper/2203.04910/) (2022)
- [EMOGI: Efficient Memory-access for Out-of-memory Graph-traversal In GPUs](https://mlsanity.com/paper/2006.06890/) (2020)

## Built on by (most cited first, 26 of 26)

- [LMCache: An Efficient KV Cache Layer for Enterprise-Scale LLM Inference](https://mlsanity.com/paper/2510.09665/) (2025)
- [Parallelizing Tool Execution and LLM Generation for Low-Latency Agent Serving](https://mlsanity.com/paper/2603.18897/) (2026)
- [ThunderAgent: A Simple, Fast and Program-Aware Agentic Inference System](https://mlsanity.com/paper/2602.13692/) (2026)
- [EVICPRESS: Joint KV-Cache Compression and Eviction for Efficient LLM Serving](https://mlsanity.com/paper/2512.14946/) (2025)
- [ContextPilot: Fast Long-Context Inference via Context Reuse](https://mlsanity.com/paper/2511.03475/) (2025)
- [Pythia: Exploiting Workflow Predictability for Efficient Agent-Native LLM Serving](https://mlsanity.com/paper/2604.25899/) (2026)
- [KVDrive: A Holistic Multi-Tier KV Cache Management System for Long-Context LLM Inference](https://mlsanity.com/paper/2605.18071/) (2026)
- [PCR: A Prefetch-Enhanced Cache Reuse System for Low-Latency RAG Serving](https://mlsanity.com/paper/2603.23049/) (2026)
- [PEEK: Context Map as an Orientation Cache for Long-Context LLM Agents](https://mlsanity.com/paper/2605.19932/) (2026)
- [Bole: Efficient Tree Speculation for Hybrid-Attention Language Models](https://mlsanity.com/paper/2608.01651/) (2026)
- [HiSparse: Scaling Sparse-Attention Decoding with Hierarchical KV Cache Management](https://mlsanity.com/paper/2608.07009/) (2026)
- [RepetitionCurse: Measuring and Understanding Router Imbalance in Mixture-of-Experts LLMs under DoS Stress](https://mlsanity.com/paper/2512.23995/) (2025)
- [ForkKV: Scaling Multi-LoRA Agent Serving via Copy-on-Write Disaggregated KV Cache](https://mlsanity.com/paper/2604.06370/) (2026)
- [SLO-Aware Compute Resource Allocation for Prefill-Decode Disaggregated LLM Inference](https://mlsanity.com/paper/2603.04716/) (2026)
- [DWDP: Distributed Weight Data Parallelism for High-Performance LLM Inference on NVL72](https://mlsanity.com/paper/2604.01621/) (2026)
- [CacheFlow: Efficient LLM Serving with 3D-Parallel KV Cache Restoration](https://mlsanity.com/paper/2604.25080/) (2026)
- [Diagnose Before You Compress: Prediction-Independent Bottleneck Witness Refinement for LLM Serving Traces](https://mlsanity.com/paper/2608.00423/) (2026)
- [vToken: Token-Level Virtualization for Reclaimable KV Caches](https://mlsanity.com/paper/2608.13263/) (2026)
- [ProServe: Unified Multi-Priority Request Scheduling for LLM Serving](https://mlsanity.com/paper/2512.12928/) (2025)
- [Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention](https://mlsanity.com/paper/2608.03555/) (2026)
- [Bounded-State Restoration: Decoupling Local Restore Capacity from External LLM State](https://mlsanity.com/paper/2608.17826/) (2026)
- [SPLASH: Co-Designing Sparse Attention with High-Bandwidth Flash for Efficient Long-Context Inference](https://mlsanity.com/paper/2609.23816/) (2026)
- [When Fancy Eviction Fails: Rethinking Cache Replacement For LLM Prefix Reuse](https://mlsanity.com/paper/2609.28870/) (2026)
- [Dynamic Flow, Static Graph: KV Cache Reuse for Efficient LLM Serving on Mobile NPUs](https://mlsanity.com/paper/2609.34727/) (2026)
- [TempoKV: Timely Staging of LLM KV Caches for Memory-Semantic Flash](https://mlsanity.com/paper/2609.35065/) (2026)
- [Adaptive DHiSparse: Speculative Decoding over a Load-Adaptive KV Hierarchy](https://mlsanity.com/paper/2610.13708/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) with expanding context windows face significant performance hurdles. While caching key-value (KV) states is critical for avoiding redundant computation, the storage footprint of long-context caches quickly exceeds GPU memory capacity, forcing production systems to adopt hierarchical caching across memory hierarchies. However, transferring large cached contexts back to the GPU introduces severe performance bottlenecks: fragmented I/O from paged layouts prevents full bandwidth utilization, and existing schedulers fail to account for cache-loading delays, leaving systems loading-bound rather than compute-bound. We present Strata, a hierarchical context caching framework designed for efficient long context LLM serving. Strata introduces GPU-assisted I/O to combat KV cache fragmentation, decoupling GPU and CPU memory layouts and employs cache-aware request scheduling to balance compute with I/O latency and overlapping unavoidable stalls with complementary tasks. Built on SGLang and deployed in production, Strata achieves up to 5x lower Time-To-First-Token (TTFT) compared to vLLM + LMCache and 3.75x speedup over NVIDIA TensorRT-LLM on long-context benchmarks, without degrading short-context performance.