GPU Integrated NIC Shared Frame Buffer Cloud Gaming Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud gaming experiences high latency due to the bottleneck of copying and transferring GPU-rendered data through CPU and network card, which increases processing time and reduces real-time streaming efficiency.
Innovation Solution
Integration of a Network Interface Controller (NIC) within the Graphics Processing Unit (GPU) allows for direct packetization and depacketization of video game content, bypassing the need for CPU involvement in data transfer, thereby reducing latency by enabling shared frame buffer access and efficient media streaming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If discrete graphics cards with GPU and separate network card are used, then video content can be rendered and streamed, but latency increases due to data copying through CPU and main memory
Solution Approach 1:
The patent integrates the network interface controller directly into the GPU architecture, merging previously separate components (GPU and network card) into a unified structure. This allows encoded video data to be transferred directly from the GPU's frame buffer to the network interface without requiring CPU intervention or main memory copying, thereby reducing latency and simplifying the data transfer path.
2Productivity
If CPU and main memory are used for data transfer, then data can be moved between GPU and network card, but processing time increases due to the bottleneck
Solution Approach 1:
The patent extracts the network interface controller from the separate network card and integrates it directly into the GPU architecture. This extraction from the traditional CPU-mediated data path eliminates the bottleneck where data must be copied to main memory and processed by the CPU, allowing direct peer-to-peer data transfer between the GPU frame buffer and network interface, thus improving streaming efficiency and reducing processing time.
3Reliability
If separate GPU and network card are used, then system modularity is maintained, but real-time streaming performance deteriorates due to data copying overhead
Solution Approach 1:
The patent merges the network interface controller with the GPU into an integrated architecture. This integration eliminates the data copying overhead associated with separate components, as encoded video data can be transferred directly from the GPU's frame buffer to the network interface through a streamlined internal pathway, significantly improving real-time streaming performance and reducing data transfer time.
Data Source
Figure 1
Figure 1a
Figure 1b
AI summary
Methods and apparatus for cloud gaming Graphics Processing Unit (GPU) with integrated Network Interface Controller (NIC) and shared frame buffer access. The GPU include one or more frame buffers that provide shared access to an integrated encoder/decoder. The GPU further includes an integrated NIC coupled to the integrated encoder/decoder and one or more video outputs coupled to the one or more frame buffers. The GPU is configured to process outbound and inbound game image content that is encoded and decoded using a video codec or using a game tile encoder and decoder. Video frames buffered in the frame buffer(s) are encoded by the integrated encoder and forwarded directly to the NIC to be packetized and streamed using a media streaming protocol. Inbound streamed media content is depacketized by the NIC and decoded by the integrated decoder, which writes the decoded content to a frame buffer to regenerate.