# KunServe: Parameter-centric Memory Management for Efficient Memory Overloading Handling in LLM Serving

arXiv: https://arxiv.org/abs/2412.18169 · HTML: https://mlsanity.com/paper/2412.18169/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Serving systems and request scheduling, one of 12 directions in Efficient LLM inference and serving.
- Builds on 24 of its 47 references that are in the map.
- First posted on arXiv 24 Dec 2024.

## Placed in

- Efficient LLM inference and serving / [Serving systems and request scheduling](https://mlsanity.com/inference/serving-systems-and-request-scheduling/)

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

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints](https://mlsanity.com/paper/2305.13245/) (2023)
- [SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot](https://mlsanity.com/paper/2301.00774/) (2023)
- [DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Serving](https://mlsanity.com/paper/2401.09670/) (2024)
- [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)
- [Mooncake: A KVCache-centric Disaggregated Architecture for LLM Serving](https://mlsanity.com/paper/2407.00079/) (2024)
- [SARATHI: Efficient LLM Inference by Piggybacking Decodes with Chunked Prefills](https://mlsanity.com/paper/2308.16369/) (2023)
- [Llumnix: Dynamic Scheduling for Large Language Model Serving](https://mlsanity.com/paper/2406.03243/) (2024)
- [Multimodal Web Navigation with Instruction-Finetuned Foundation Models](https://mlsanity.com/paper/2305.11854/) (2023)
- [Fast Distributed Inference Serving for Large Language Models](https://mlsanity.com/paper/2305.05920/) (2023)
- [Inference without Interference: Disaggregate LLM Inference for Mixed Downstream Workloads](https://mlsanity.com/paper/2401.11181/) (2024)
- [LoongServe: Efficiently Serving Long-Context Large Language Models with Elastic Sequence Parallelism](https://mlsanity.com/paper/2404.09526/) (2024)
- [DeepSpeed-FastGen: High-throughput Text Generation for LLMs via MII and DeepSpeed-Inference](https://mlsanity.com/paper/2401.08671/) (2024)
- [vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention](https://mlsanity.com/paper/2405.04437/) (2024)
- [SpotServe: Serving Generative Large Language Models on Preemptible Instances](https://mlsanity.com/paper/2311.15566/) (2023)
- [ServerlessLLM: Low-Latency Serverless Inference for Large Language Models](https://mlsanity.com/paper/2401.14351/) (2024)
- [Evaluating Quantized Large Language Models](https://mlsanity.com/paper/2402.18158/) (2024)
- [POD-Attention: Unlocking Full Prefill-Decode Overlap for Faster LLM Inference](https://mlsanity.com/paper/2410.18038/) (2024)
- [How Does Quantization Affect Multilingual LLMs?](https://mlsanity.com/paper/2407.03211/) (2024)
- [InferCept: Efficient Intercept Support for Augmented Large Language Model Inference](https://mlsanity.com/paper/2402.01869/) (2024)

## Built on by (most cited first, 11 of 11)

- [WarmServe: Enabling One-for-Many GPU Prewarming for Multi-LLM Serving](https://mlsanity.com/paper/2512.09472/) (2025)
- [Fast LLM Post-training via Decoupled and Fastest-of-N Speculation](https://mlsanity.com/paper/2511.16193/) (2025)
- [AGFT: An Adaptive GPU Frequency Tuner for Real-Time LLM Inference Optimization](https://mlsanity.com/paper/2508.01744/) (2025)
- [SMetric: Rethink LLM Scheduling for Serving Agents with Balanced Session-centric Scheduling](https://mlsanity.com/paper/2607.08565/) (2026)
- [Autopoiesis: A Self-Evolving System Paradigm for LLM Serving Under Runtime Dynamics](https://mlsanity.com/paper/2604.07144/) (2026)
- [Scepsy: Serving Agentic Workflows Using Aggregate LLM Pipelines](https://mlsanity.com/paper/2604.15186/) (2026)
- [ObjectCache: Layerwise Object-Storage Retrieval for KV Cache Reuse](https://mlsanity.com/paper/2605.22850/) (2026)
- [Crossflow: Prefill-Decode Elasticity for Agentic LLM Serving](https://mlsanity.com/paper/2609.27085/) (2026)
- [ServerlessT2I: Efficient Text-to-Image Workflow Serving on a Serverless Platform](https://mlsanity.com/paper/2607.26566/) (2026)
- [Dynamic HBM Repartitioning for Multi-Turn MoE Serving](https://mlsanity.com/paper/2609.13537/) (2026)
- [FastEEP: Fast Intra-Instance Elastic Scaling for Expert-Parallel MoE Serving](https://mlsanity.com/paper/2610.18451/) (2026)

## Abstract (arXiv)

Serving LLMs with a cluster of GPUs is common nowadays, where the serving system must meet strict latency SLOs required by applications. However, the stateful nature of LLM serving requires maintaining huge states (i.e., KVCache) in limited GPU memory. Under spikes in real-world workloads, GPU memory can be easily throttled, leading to orders of magnitude higher response latency due to queuing introduced by waiting for KVCache to be reclaimed. Prior KVCache-centric approaches handle load throttling by dropping, migrating, or swapping KVCache. These methods fail to release sufficient memory quickly with requests still queued. This paper proposes the first parameter-centric approach to handling throttling by selectively dropping replicated parameters to instantly free memory for requests, based on an unnoticed observation that model parameters are commonly replicated across GPUs for serving LLMs. With additional memory, all requests can be served with a larger batch without queuing. To make the parameter-centric approach correct and efficient, we cooperatively execute requests on GPUs with a complete copy of parameters using pipeline parallelism, and derive an appropriate drop plan without unnecessary cooperation. We also design techniques to minimize the performance overhead due to pipeline parallelism with the execution patterns of requests under drop. Evaluations show that sys reduces the tail TTFT of requests under throttling by up to 72.2 times compared to the state-of-the-art systems including Llumnix, vLLM and InferCept.