GPU Emulator Tile-Based Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud-based GPU emulation faces inefficiencies and latency due to uneven processing load balancing during rasterization, particularly in software emulations, which can lead to increased resource demands and high costs during peak usage.
Innovation Solution
Implementing a tile-based rasterization method where a virtual image is divided into tiles, with each tile having a predetermined number of pixels, allowing for efficient load balancing by distributing processing between rasterization threads through a load balancer, ensuring each thread renders a similar number of pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional rasterization is used without tile-based division, then the processing load is not balanced, but the system complexity is reduced
Solution Approach 1:
The patent applies segmentation by dividing the virtual image into multiple tiles, each with a predetermined number of pixels. This allows the rasterization process to be segmented into manageable units that can be distributed across multiple threads, achieving load balancing while maintaining manageable system complexity through structured division rather than unstructured complexity
2Productivity
If more rasterization threads are dedicated to processing, then the processing capacity increases, but the resource consumption increases
Solution Approach 1:
The patent implements dynamics by making the number of active rasterization threads adjustable rather than fixed. The system can dynamically allocate threads based on the actual processing load identified through tile analysis, allowing the system to scale resource consumption up or down to match productivity needs, avoiding both over-provisioning and under-provisioning of resources
3Use of energy by moving object
If the frame rate is decreased to reduce processing load, then the resource consumption decreases, but the visual quality deteriorates
Solution Approach 1:
The patent applies local quality by differentiating the processing requirements of different tiles based on their content. Not all tiles require the same processing intensity, so the system can allocate resources locally to tiles that need them most, maintaining high frame rates for critical regions while reducing overall resource consumption through selective processing of less demanding areas
Data Source
AI summary
Aspects of the present disclosure describe a software based emulator of a graphics processing unit (GPU) that is configured to operate over a cloud-based network. A virtual image containing graphics primitives is divided into a plurality of tiles. A load balancer assigns tiles to rasterization threads in order to evenly distribute the processing load. The rasterization threads then rasterize their assigned tiles and deliver rendered pixels to a frame buffer. The frame buffer builds a frame from the rendered pixels and then delivers the frame over the network to a client device platform. It is emphasized that this abstract is provided to comply with the rules requiring an abstract that will allow a searcher or other reader to quickly ascertain the subject matter of the technical disclosure. It is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims.