GPU Database Analytics Processor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional approaches to processing unstructured data are inefficient due to the lack of effective field indexing and computational intensity, making it difficult to analyze and extract results from large volumes of unstructured data, especially in the context of Big Data analytics which requires scalable and cost-effective solutions.
Innovation Solution
Employing Graphical Processing Units (GPUs) in conjunction with solid-state storage to transform unstructured data into a delimiterless, columnar form, enabling parallel processing and efficient matching operations, thereby alleviating IO bottlenecks and reducing the need for large clusters of computers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional field indexing and computed field positions are used in unstructured databases, then data storage flexibility is maintained, but query processing efficiency deteriorates due to the need to traverse large ranges of memory and perform repetitive matching operations
Solution Approach 1:
The patent applies preliminary action by pre-processing unstructured data into a delimiterless columnar form before query operations. This transformation organizes data into sequential strings with field definition characters removed, creating a structure that enables efficient GPU processing without sacrificing the flexibility of unstructured data storage. The data is prepared in advance for parallel processing operations.
Solution Approach 2:
The patent substitutes mechanical CPU-based sequential processing with GPU-based parallel processing. By leveraging the massive parallelism of GPUs, the system performs matching and comparison operations simultaneously across multiple data elements, dramatically improving query processing efficiency while maintaining adaptability to unstructured data formats.
2Power
If large clusters of computers are used to process Big Data volumes, then computational power is sufficient, but system complexity and cost increase significantly
Solution Approach 1:
The patent merges multiple computational functions into a unified GPU processing architecture. By combining data transformation, parallel processing, and result aggregation capabilities within the GPU system, the patent achieves high computational power with reduced system complexity compared to distributed computer clusters.
Solution Approach 2:
The patent employs GPUs as universal processing units that can handle multiple types of operations including data transformation, parallel matching, comparison operations, and result generation. This multi-functionality eliminates the need for specialized hardware components and complex system architectures required by conventional approaches.
3Device complexity
If conventional CPU-based processing is used for unstructured data, then system simplicity is maintained, but processing speed deteriorates due to computational intensity and repetitive operations
Solution Approach 1:
The patent transitions from single-threaded CPU processing to multi-threaded GPU processing by exploiting the parallel dimension. GPUs provide thousands of cores that can execute operations simultaneously, adding a parallel processing dimension that dramatically increases speed while maintaining architectural simplicity through standardized hardware interfaces.
Data Source
AI summary
An analytics processing system generates analytics from a collection of unstructured data by. transforming a received source of input data from an unstructured database into a delimiterless form, and iteratively moving portions of the delimiterlesss input data from a solid-state memory to a shared memory adapted for parallel operations with a plurality of GPU cores. The method stores computational data, such as values for matching, in a high speed memory responsive to operations with the shared memory, in which the high-speed memory remains static for the duration of the iterations. A host CPU invokes the plurality of cores for performing the parallel operations on the computational data and the portions of the delimiterless input data, and stores a result in a general memory accessible from a graphical user interface (GUI). The GPU cores parallelize the matching task of the input data from the unstructured database against the match data.