- Technology
- Computer Science
- Distributed Systems
- High Performance Computing
- Parallel Computing
- GPU Programming
- CUDA
- Technology
- Cloud Computing
- Amazon Web Services (AWS)
- AWS Networking & Content Delivery
- Amazon Elastic Load Balancer
- Technology
- Computer Science
- Distributed Systems
- High Performance Computing
- Parallel Computing
- GPU Computing
gVulkan - Scalable GPU Pooling for Pixel-Grained Rendering in Ray Tracing
AI, Data Science & Cloud Certificates from Google, IBM & Meta
Finance Certifications Goldman Sachs and Amazon Teams Trust
Overview
Google, IBM & Meta Certificates – 40% Off
One plan covers every Professional Certificate on Coursera.
Unlock All Certificates
Explore a groundbreaking conference talk on gVulkan, the first transparent multi-GPU acceleration rendering solution for Vulkan-based ray tracing. Delve into how this innovative technology addresses the computational demands of ray tracing rendering by offloading to the cloud via API-forwarding. Learn about the implementation of Split Frame Rendering (SFR) that enables parallel acceleration across multiple GPUs, dynamically rebalancing workload at a pixel-grained level. Discover the impressive performance gains achieved by gVulkan, including a 3.81× speedup across 4 GPUs on average. Gain insights into how this solution can enhance scene realism and provide immersive experiences while overcoming the limitations of traditional local GPU rendering.
Syllabus
USENIX ATC '24 - gVulkan: Scalable GPU Pooling for Pixel-Grained Rendering in Ray Tracing
Taught by
USENIX