# TileLink: Generating Efficient Compute-Communication Overlapping Kernels using Tile-Centric Primitives

arXiv: https://arxiv.org/abs/2503.20313 · HTML: https://mlsanity.com/paper/2503.20313/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Mar 2025.

## Built on by (most cited first, 24 of 24)

- [Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures](https://mlsanity.com/paper/2505.09343/) (2025)
- [MegaScale-MoE: Large-Scale Communication-Efficient Training of Mixture-of-Experts Models in Production](https://mlsanity.com/paper/2505.11432/) (2025)
- [Robix: A Unified Model for Robot Interaction, Reasoning and Planning](https://mlsanity.com/paper/2509.01106/) (2025)
- [LLM-42: Enabling Determinism in LLM Inference with Verified Speculation](https://mlsanity.com/paper/2601.17768/) (2026)
- [TokenWeave: Efficient Compute-Communication Overlap for Distributed LLM Inference](https://mlsanity.com/paper/2505.11329/) (2025)
- [Efficient and Adaptable Overlapping for Computation and Communication via Signaling and Reordering](https://mlsanity.com/paper/2504.19519/) (2025)
- [UniEP: Unified Expert-Parallel MoE MegaKernel for LLM Training](https://mlsanity.com/paper/2604.19241/) (2026)
- [Kareus: Joint Reduction of Dynamic and Static Energy in Large Model Training](https://mlsanity.com/paper/2601.17654/) (2026)
- [Axe: A Simple Unified Layout Abstraction for Machine Learning Compilers](https://mlsanity.com/paper/2601.19092/) (2026)
- [ZeCO: Zero Communication Overhead Sequence Parallelism for Linear Attention](https://mlsanity.com/paper/2507.01004/) (2025)
- [FlexLink: Boosting your NVLink Bandwidth by 27% without accuracy concern](https://mlsanity.com/paper/2510.15882/) (2025)
- [SwiftFusion: Scalable Sequence Parallelism for Distributed Inference of Diffusion Transformers on GPUs](https://mlsanity.com/paper/2601.20273/) (2026)
- [CUCo: An Agentic Framework for Compute and Communication Co-design](https://mlsanity.com/paper/2603.02376/) (2026)
- [Design Space Exploration of DMA based Finer-Grain Compute Communication Overlap](https://mlsanity.com/paper/2512.10236/) (2025)
- [Stage-Aware Communication Scheduling for Disaggregated LLM Serving](https://mlsanity.com/paper/2603.17456/) (2026)
- [Ada-MK: Adaptive MegaKernel Optimization via Automated DAG-based Search for LLM Inference](https://mlsanity.com/paper/2605.11581/) (2026)
- [Searching for Plans You Can Actually Build: A Realizability-Aware Full-Space Optimizer for MoE Training and Serving](https://mlsanity.com/paper/2607.18631/) (2026)
- [Fine-grained Computation-Communication Overlap via Tile-level Signaling and Scheduling for Mixture-of-Experts](https://mlsanity.com/paper/2607.19539/) (2026)
- [CommFuse: Hiding Tail Latency via Communication Decomposition and Fusion for Distributed LLM Training](https://mlsanity.com/paper/2604.24013/) (2026)
- [DynaFlow: Transparent and Flexible Intra-Device Parallelism via Programmable Operator Scheduling](https://mlsanity.com/paper/2605.21603/) (2026)
- [Resource-aware Computation-Communication Overlap for multi-GPU ML Workloads](https://mlsanity.com/paper/2606.09200/) (2026)
- [X-Stage: Modeling Post-Issue Backpressure in GPU Communication--Computation Fusion](https://mlsanity.com/paper/2607.23264/) (2026)
- [Analytical Resource Management for Fine-grained MoE Computation-Communication Overlap](https://mlsanity.com/paper/2609.07536/) (2026)
- [mKernel: Fast Multi-GPU, Multi-Node Fused Kernels](https://mlsanity.com/paper/2609.13585/) (2026)

## Abstract (arXiv)

Large deep learning models have achieved state-of-the-art performance in a wide range of tasks. These models often necessitate distributed systems for efficient training and inference. The fundamental building blocks for distributed model execution are intra-layer parallel operators. The most effective approach to enhancing the performance of intra-layer parallel operators involves overlapping computation with communication. The overlapping can be achieved through either operator decomposition or kernel fusion. While decomposing operators is straightforward to implement, it often results in suboptimal performance. On the other hand, fusing communication kernels with compute kernels demands significant expertise and is error-prone. In this paper, we propose TileLink to enable efficient compilation and generation of overlapped compute-communication kernels. TileLink is composed of frontend and backend. In the frontend, TileLink decouples the design space of communication and computation, linking these two parts via tile-centric primitives. In the backend, TileLink translates these primitives into low-level communication instructions, integrating the communication and computation components to achieve overlapped execution. In experiments, TileLink achieves from 1.17× to 20.76× speedup to non-overlapping baseline and achieves performance comparable to state-of-the-art overlapping libraries on GPUs.