Seismic Co-Processor Data Compression for I/O Bottlenecks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional configurations of host computing systems with co-processor devices for seismic data processing face bottlenecks due to communication limitations between the host and co-processor devices, as well as within the co-processor devices themselves, which hinder efficient processing of seismic data.
Innovation Solution
A method and system where compressed input data is received by a co-processor device, stored, and processed using a co-processor, with the ability to decompress and recompress data as needed, allowing for efficient data transfer and processing, including the use of multiple co-processor devices interconnected for parallel processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data is transmitted in uncompressed format between host computing system and co-processor devices, then data processing accuracy is maintained, but communication bandwidth requirements increase and processing speed decreases
Solution Approach 1:
The patent applies data compression to change the physical state of data from uncompressed to compressed form, reducing the volume of data transmitted between host computing systems and co-processor devices. This parameter change directly addresses the contradiction by decreasing communication bandwidth requirements and improving processing speed without sacrificing data integrity, as the compressed data can be decompressed to its original form for accurate processing
2Productivity
If co-processor devices store large amounts of seismic data, then processing capacity increases, but storage memory requirements exceed available resources
Solution Approach 1:
The patent implements compression of seismic data to fundamentally change the storage parameter, allowing co-processor devices to store significantly more data within the same physical memory constraints. By compressing data before storage and decompressing when needed, the system achieves enhanced processing capacity without requiring additional hardware memory resources
3Loss of information
If communication between host computing system and co-processor devices is increased for data transfer, then data availability improves, but communication bottlenecks worsen
Solution Approach 1:
The patent applies compression to change the transmission parameter of data, reducing the amount of data that needs to be communicated between host computing systems and co-processor devices. This parameter change directly reduces communication time and eliminates bottlenecks while maintaining complete data availability, as compressed data can be fully decompressed to retrieve all original information
Data Source
AI summary
A system and method for processing seismic data on one or more co-processor devices that are operatively coupled to a host computing system via a communications channel. The compression of input data transmitted to the co-processor device and/or the size of the storage provided on the co-processor device may enhance the efficiency of the processing of the data on the peripheral device by obviating a bottleneck caused by the relatively slow transfer of data between the host computing system and the co-processor device or by the relatively slow transfer of data within the co-processor device between the co-processor information storage and the co-processor.