Asplos 23 Session 8c Ugrapher High Performance Graph Operator
![asplos 23 Session 8c Ugrapher High Performance Graph Operator asplos 23 Session 8c Ugrapher High Performance Graph Operator](https://ytimg.googleusercontent.com/vi/lmkbJDChD9U/hqdefault.jpg?sqp=-oaymwEmCOADEOgC8quKqQMa8AEB-AHSBoAC4AOKAgwIABABGGUgWChlMA8=&rs=AOn4CLAfVjRvnT4D7vKeOyKaw5OPdgRTuw)
Asplos 23 Session 8c Ugrapher High Performance Graph Operator In this work, we propose ugrapher, a unified interface that achieves general high performance for different graph operators and datasets. the existing gnn frameworks can easily integrate our design for its simple and unified api. we take a principled approach that decouples a graph operator’s computation and schedule to achieve that. Asplos'23: the 28th international conference on architectural support for programming languages and operating systemssession 8c: graph bsession chair: tony.
![Asplos 23 系统顶会论文 Plugsched 安全 高效的多场景调度器热升级详解 知乎 Asplos 23 系统顶会论文 Plugsched 安全 高效的多场景调度器热升级详解 知乎](https://pic4.zhimg.com/v2-959c40bdc6af09d3c48709193b1807d3_b.jpg)
Asplos 23 系统顶会论文 Plugsched 安全 高效的多场景调度器热升级详解 知乎 Ugrapher: high performance graph operator computation via unified abstraction for graph neural networks. published in asplos 23, 2022. paper. share on twitter facebook linkedin previous next. This work proposes ugrapher, a unified interface that achieves general high performance for different graph operators and datasets and explores various schedule strategies based on the abstraction that can balance the well established trade off relationship between parallelism, locality, and efficiency. as graph neural networks (gnns) have achieved great success in many graph learning problems. Wu. 2023. ugrapher: high performance graph operator computation via unified abstraction for graph neural networks. inproceedings of the 28th acm international conference on architectural support for programming languages and operating systems, volume 2 (asplos ’23), march 25–29, 2023, vancouver, bc, canada. acm, new york, ny, usa,14pages.https:. Gunrock achieves a balance between performance and expressiveness by coupling high performance gpu computing primitives and optimization strategies with a high level programming model that allows.
人工智能 Asplos 2023 图神经网络统一图算子抽象 Ugrapher 大幅提高计算性能 个人文章 Segmentfault 思否 Wu. 2023. ugrapher: high performance graph operator computation via unified abstraction for graph neural networks. inproceedings of the 28th acm international conference on architectural support for programming languages and operating systems, volume 2 (asplos ’23), march 25–29, 2023, vancouver, bc, canada. acm, new york, ny, usa,14pages.https:. Gunrock achieves a balance between performance and expressiveness by coupling high performance gpu computing primitives and optimization strategies with a high level programming model that allows. In this work, we propose ugrapher, a unified interface that achieves general high performance for different graph operators and datasets. the existing gnn frameworks can easily integrate our design for its simple and unified api. we take a principled approach that decouples a graph operator’s computation and schedule to achieve that. 28th acm international conference on architectural support for programming languages and operating systems, volume 2 (asplos 2023), march 25–29, 2023, vancouver, bc, canada. asplos 2023 – proceedings. contents abstracts authors.
Comments are closed.