• [Experience sharing] GPU CUDA uses memory padding to avoid bank conflict


      Welcome to pay attention to my public account [Jizhi Vision], reply 001 to get Google programming specifications   O_o   >_<   o_O   O_o   ~_~   o_O This article talks about how to use memory padding in GPU CUDA programming to avoid bank conflict. 1、Shared memory Shared memory is a small, low-latency on-chip […]

  • Comparison between WebGL and WebGPU [8] – Summary and thoughts at the end of the series


    Postscript to the end of the series Lin Lin always wrote some articles to compare the conventional functions of WebGL and WebGPU. Maybe he is a bit amateurish in understanding the API, and even does not rule out that there are mistakes in the expression, but I hope that the correct part will inspire readers […]

  • Comparison of WebGL and WebGPU [7] – the destination of rendering


    1. Overview In fact, the comparison of the basic conventional APIs in the sixth article is almost finished (except for GPGPU and query APIs that are not involved), but there is one detail that is still worth thinking about and recording in a comparison article. is where to render. The context object of WebGL is […]

  • Milvus query task scheduling principle


    This article mainly explains how Milvus schedules query tasks in a single-machine multi-card scenario. In addition, we will also discuss the problems encountered during the scheduling implementation, solutions and future development directions. background As mentioned in Data Management in the Large-Scale Vector Retrieval Scenario (Part 1), the similarity search of vectors is actually judged by […]

  • Gpushare.com | How to use TensorBoardX visualization tool?


    TensorBoardX is based on TensorBoard, a tool that can be used for Pytorch data visualization. Users who are familiar with TensorBoard can also easily get started with TensorBoardX~ Let’s take a look at how to use TensorBoardX on our platform? Install TensorBoardX This demonstration uses TensorBoardX visualization for the Pytorch framework, creates an instance of […]