High Performance Computing in web application
0 likes143 views
This document discusses high performance computing techniques for web applications, including WebGL, WebGPU, and WebAssembly. WebGPU aims to standardize GPU access in the browser by mapping to native APIs like Vulkan and Direct3D. It uses pipelines and a binding model to improve performance. The author proposes investigating WebGPU further to improve web VR applications and determine if it can help run programs across multiple GPUs for even better performance.
High Performance Computing in web application
- 1. High Performance Computing in Web application Rëza Habibi
- 2. Outlines 1- Quick Review 2- How web graphic works? 3- WebGL 4- WebGPU 4- WebAssembly 5- Web Big applications 6- Proposed Idea 2
- 3. A Quick Review 3 • Graphic Process Unit • WebGPU • WASM • Applications
- 4. 4 How web graphic works?
- 5. 5 WebGL • It draws points, lines, and triangles based on code you supply
- 6. 6 WebGL • WebGL only cares about 2 things: Clip space coordinates and Colors Vertex Shader Example :
- 7. 7 WebGL • WebGL only cares about 2 things: Clip space coordinates and Colors Fragment shader Example :
- 8. WebAssembly 8 • WebAssembly (WASM )is a safe, portable, low-level code format designed for efficient execution and compact representation. Its main goal is to enable high performance applications on the Web, but it does not make any Web-specific assumptions or provide Web- specific features, so it can be employed in other environments as well. • It is Fast, safe, and portable as well as
- 9. WebGPU 9 • It designed to map to the VulKan, Direct3D 12 , and Metal Native GPU APIs. • Vulkan , Direct 3D , Metal?
- 10. WebGPU 10
- 11. WebGPU PIPLine 11 Create one pipeline object one call Create one resource , one call
- 13. 1313 Project Questions and Challenges? • Q1 : Does WebGPU improve the performance of web applications ? • Q2 : Can we run our program on several GPU rather than one GPU?
- 14. Why? 14
- 15. Web VR Web Big Application 15 Example of Web big application: NeuroGlancer Web VR
- 16. Proposed Idea 16 • Improving Web Virtual Reality application via WebGPU. Using GPU power for answering the two questions and investigate whether we can solve these challenges or not. I’ll use GPUPiplines and GPUbinding Model for solving mentioned challenges.
- 17. 17 Thank you. 17