Asynchronous Compute for Large-Scale Data Rendering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Large-scale raw data sets require extensive preprocessing, which can take hours or days and may introduce unwanted filtering effects, leading to misleading analysis results in user-interactive environments.

Innovation Solution

An asynchronous computing and rendering system that organizes data into subregions and utilizes parallel computing and rendering clusters with a data interpretation platform for immediate visualization, hiding network transfer costs and latency to provide real-time rendering capabilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If large-scale raw data is preprocessed to create a reduced data set, then the data can be processed and visualized in user-interactive environments, but the preprocessing time can take hours or even days and may introduce unwanted filtering effects

Engineering Contradiction:

Improvedata processing speedVSAvoidpreprocessing time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent divides the large-scale data set into multiple data subregions and processes them in parallel across multiple computing machines. Each computing machine processes a subset of subregions independently, reducing the total preprocessing time from hours/days to a fraction of that time while maintaining data integrity through the subregion partitioning approach

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary data organization by dividing raw data into subregions before the actual processing begins. This preliminary segmentation enables subsequent parallel processing to proceed much faster without introducing filtering effects, as the data structure is prepared in advance but not processed until needed

Inventive Principle:

Principle #10Preliminary action

2Loss of time

If data is organized into subregions and processed in parallel, then preprocessing time is reduced significantly, but the system complexity increases with multiple computing and rendering clusters

Engineering Contradiction:

Improvepreprocessing timeVSAvoidsystem architecture complexity

Core Design Contradiction:

Loss of timeVSDevice complexity

Solution Approach 1:

The patent introduces a data interpretation platform as an intermediary layer between the raw data storage and the rendering cluster. This intermediary manages the complex coordination between multiple computing machines and rendering machines, handling data distribution, compute result collection, and rendering coordination, thereby reducing the operational complexity despite the parallel architecture

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system merges the computing cluster and rendering cluster into a unified asynchronous computing and rendering system where compute results are directly fed to rendering machines. This integration eliminates the need for separate data transfer and processing stages, reducing overall system complexity while maintaining parallel processing capabilities

Inventive Principle:

Principle #5Merging (Combining)

3Measurement precision

If traditional preprocessing is used, then data can be prepared for analysis, but network transfer costs and latency are not hidden, resulting in slower real-time visualization

Engineering Contradiction:

Improvedata analysis accuracyVSAvoidreal-time visualization speed

Core Design Contradiction:

Measurement precisionVSSpeed

Solution Approach 1:

The patent implements continuous asynchronous processing where data is continuously divided into subregions and processed in parallel while waiting for user interaction. The system maintains ready-state rendering by continuously processing subregions in the background, eliminating idle waiting time and achieving real-time visualization speeds without sacrificing data analysis accuracy

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The system performs preliminary compute result generation for data subregions in advance, but only processes and renders them when needed based on user interaction. This preliminary preparation of data structures without immediate processing allows the system to respond instantly to user actions while maintaining data integrity for accurate analysis

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9117284B2Asynchronous compute integrated into large-scale data rendering using dedicated, separate computing and rendering clusters

Publication Date: 2015.08.25 NVIDIA CORP

AI summary

An asynchronous computing and rendering system includes a data storage unit that provides storage for processing a large-scale data set organized in accordance to data subregions and a computing cluster containing a parallel plurality of asynchronous computing machines that provide compute results based on the data subregions. The asynchronous computing and rendering system also includes a rendering cluster containing a parallel multiplicity of asynchronous rendering machines coupled to the asynchronous computing machines, wherein each rendering machine renders a subset of the data subregions. Additionally, the asynchronous computing and rendering system includes a data interpretation platform coupled to the asynchronous rendering machines that provides user interaction and rendered viewing capabilities for the large-scale data set. An asynchronous computing and rendering method is also provided.