# Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention

arXiv: https://arxiv.org/abs/2608.03555 · HTML: https://mlsanity.com/paper/2608.03555/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in KV cache optimization, one of 12 directions in Efficient LLM inference and serving.
- Builds on 48 of its 96 references that are in the map.
- First posted on arXiv 4 Aug 2026.

## Placed in

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

## Builds on (most cited first, 30 of 48)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [SGLang: Efficient Execution of Structured Language Model Programs](https://mlsanity.com/paper/2312.07104/) (2023)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models](https://mlsanity.com/paper/2512.02556/) (2025)
- [DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Serving](https://mlsanity.com/paper/2401.09670/) (2024)
- [GLM-5: from Vibe Coding to Agentic Engineering](https://mlsanity.com/paper/2602.15763/) (2026)
- [Splitwise: Efficient generative LLM inference using phase splitting](https://mlsanity.com/paper/2311.18677/) (2023)
- [Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve](https://mlsanity.com/paper/2403.02310/) (2024)
- [Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference](https://mlsanity.com/paper/2406.10774/) (2024)
- [SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?](https://mlsanity.com/paper/2509.16941/) (2025)
- [MoBA: Mixture of Block Attention for Long-Context LLMs](https://mlsanity.com/paper/2502.13189/) (2025)
- [Mooncake: A KVCache-centric Disaggregated Architecture for LLM Serving](https://mlsanity.com/paper/2407.00079/) (2024)
- [SparQ Attention: Bandwidth-Efficient LLM Inference](https://mlsanity.com/paper/2312.04985/) (2023)
- [DynamoLLM: Designing LLM Inference Clusters for Performance and Energy Efficiency](https://mlsanity.com/paper/2408.00741/) (2024)
- [MagicPIG: LSH Sampling for Efficient LLM Generation](https://mlsanity.com/paper/2410.16179/) (2024)
- [Efficient Interactive LLM Serving with Proxy Model-based Sequence Length Prediction](https://mlsanity.com/paper/2404.08509/) (2024)
- [NeuPIMs: NPU-PIM Heterogeneous Acceleration for Batched LLM Inferencing](https://mlsanity.com/paper/2403.00579/) (2024)
- [ASTRA-sim2.0: Modeling Hierarchical Networks and Disaggregated Systems for Large-model Training at Scale](https://mlsanity.com/paper/2303.14006/) (2023)
- [ExeGPT: Constraint-Aware Resource Scheduling for LLM Inference](https://mlsanity.com/paper/2404.07947/) (2024)
- [MiniMax Sparse Attention](https://mlsanity.com/paper/2606.13392/) (2026)
- [SWE-chat: Coding Agent Interactions From Real Users in the Wild](https://mlsanity.com/paper/2604.20779/) (2026)
- [Duplex: A Device for Large Language Models with Mixture of Experts, Grouped Query Attention, and Continuous Batching](https://mlsanity.com/paper/2409.01141/) (2024)
- [InfLLM-V2: Dense-Sparse Switchable Attention for Seamless Short-to-Long Adaptation](https://mlsanity.com/paper/2509.24663/) (2025)
- [LLMServingSim: A HW/SW Co-Simulation Infrastructure for LLM Inference Serving at Scale](https://mlsanity.com/paper/2408.05499/) (2024)

## Abstract (arXiv)

This paper presents a heterogeneous decode-phase serving system that relocates the KV cache out of GPU memory, motivated by the retrieval-based sparse attention that recent frontier LLMs adopt to serve million-token contexts. It partitions a decode step by operation type: GPU nodes hold the model weights and execute the projections and MoE layers, while processing-near-memory (PNM) nodes hold the KV cache and index keys and execute every operation that reads them. We first show that the assumptions behind prior PIM and PNM designs no longer hold for these operations, and derive four design requirements for such a node. From these requirements, we propose KARAT (KV-cache-resident Accelerator for Retrieval-based ATtention), a general-purpose PNM design that is the design point meeting all four. A KARAT device combines large LPDDR capacity with general-purpose compute sized for the retrieval indexer, serving an operational intensity beyond what PIM/PNM designs built for low-intensity GEMV target while accommodating diverse sparse attention algorithms that fixed-function units cannot support as they evolve. To reduce pipeline bubbles as the two device types alternate between micro-batches, we further propose opportunistic, fine-grained micro-batch scheduling (OFMS), which hides expert all-to-all behind the other micro-batch's GEMMs, and context-length-aware micro-batch rebalancing (CMR), which equalizes their token counts despite the variance in context length. Across three state-of-the-art models and real agentic traces, our proposed system improves throughput per TDP under a service-level objective by 2.09-6.13x over a GPU-only baseline and runs training-free sparse attention methods with 1.36-3.21x improvements.