# FLYING SERVING: On-the-Fly Parallelism Switching for Large Language Model Serving

arXiv: https://arxiv.org/abs/2602.22593 · HTML: https://mlsanity.com/paper/2602.22593/

## Key facts

- Cited by 6 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 26 of its 51 references that are in the map.
- First posted on arXiv 26 Feb 2026.

## 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, 26 of 26)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [Competition-Level Code Generation with AlphaCode](https://mlsanity.com/paper/2203.07814/) (2022)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints](https://mlsanity.com/paper/2305.13245/) (2023)
- [Executable Code Actions Elicit Better LLM Agents](https://mlsanity.com/paper/2402.01030/) (2024)
- [Efficiently Scaling Transformer Inference](https://mlsanity.com/paper/2211.05102/) (2022)
- [Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve](https://mlsanity.com/paper/2403.02310/) (2024)
- [MLPerf Inference Benchmark](https://mlsanity.com/paper/1911.02549/) (2019)
- [SARATHI: Efficient LLM Inference by Piggybacking Decodes with Chunked Prefills](https://mlsanity.com/paper/2308.16369/) (2023)
- [Fast Distributed Inference Serving for Large Language Models](https://mlsanity.com/paper/2305.05920/) (2023)
- [LoongServe: Efficiently Serving Long-Context Large Language Models with Elastic Sequence Parallelism](https://mlsanity.com/paper/2404.09526/) (2024)
- [DynamoLLM: Designing LLM Inference Clusters for Performance and Energy Efficiency](https://mlsanity.com/paper/2408.00741/) (2024)
- [BurstGPT: A Real-world Workload Dataset to Optimize LLM Serving Systems](https://mlsanity.com/paper/2401.17644/) (2024)
- [Tenplex: Dynamic Parallelism for Deep Learning using Parallelizable Tensor Collections](https://mlsanity.com/paper/2312.05181/) (2023)
- [Seesaw: High-throughput LLM Inference via Model Re-sharding](https://mlsanity.com/paper/2503.06433/) (2025)
- [From 128K to 4M: Efficient Training of Ultra-Long Context Large Language Models](https://mlsanity.com/paper/2504.06214/) (2025)
- [TokenScale: Timely and Accurate Autoscaling for Disaggregated LLM Serving with Token Velocity](https://mlsanity.com/paper/2512.03416/) (2025)
- [Hetis: Serving LLMs in Heterogeneous GPU Clusters with Fine-grained and Dynamic Parallelism](https://mlsanity.com/paper/2509.08309/) (2025)
- [Shift Parallelism: Low-Latency, High-Throughput LLM Inference for Dynamic Workloads](https://mlsanity.com/paper/2509.16495/) (2025)
- [HAP: Hybrid Adaptive Parallelism for Efficient Mixture-of-Experts Inference](https://mlsanity.com/paper/2508.19373/) (2025)
- [Arctic Inference with Shift Parallelism: Fast and Efficient Open Source Inference System for Enterprise AI](https://mlsanity.com/paper/2507.11830/) (2025)

## Built on by (most cited first, 6 of 6)

- [Foundry: Template-Based CUDA Graph Context Materialization for Fast LLM Serving Cold Start](https://mlsanity.com/paper/2604.06664/) (2026)
- [Moebius: Serving Mixture-of-Expert Models with Seamless Runtime Parallelism Switch](https://mlsanity.com/paper/2606.26607/) (2026)
- [Crossflow: Prefill-Decode Elasticity for Agentic LLM Serving](https://mlsanity.com/paper/2609.27085/) (2026)
- [FlexMoE: One-for-All Nested Intra-Expert Pruning for MoE Language Models](https://mlsanity.com/paper/2606.27866/) (2026)
- [SPLASH: Switching Parallel Layouts of Attention with Seamless Handoff for LLM Serving](https://mlsanity.com/paper/2609.37626/) (2026)
- [Scaling LLM Inference via Parallelism: Throughput-Latency Trade-off](https://mlsanity.com/paper/2610.19405/) (2026)

## Abstract (arXiv)

Production LLM serving must simultaneously deliver high throughput, low latency, and sufficient context capacity under non-stationary traffic and mixed request requirements. Data parallelism (DP) maximizes throughput by running independent replicas, while tensor parallelism (TP) reduces per-request latency and pools memory for long-context inference. However, existing serving stacks typically commit to a static parallelism configuration at deployment; adapting to bursts, priorities, or long-context requests is often disruptive and slow. We present Flying Serving, a vLLM-based system that enables online DP-TP switching without restarting engine workers. Flying Serving makes reconfiguration practical by virtualizing the state that would otherwise force data movement: (i) a zero-copy Model Weights Manager that exposes TP shard views on demand, (ii) a KV Cache Adaptor that preserves request KV state across DP/TP layouts, (iii) an eagerly initialized Communicator Pool to amortize collective setup, and (iv) a deadlock-free scheduler that coordinates safe transitions under execution skew. Across three popular LLMs and realistic serving scenarios, Flying Serving improves performance by up to 4.79× under high load and 3.47× under low load while supporting latency- and memory-driven requests.