Video Processing In Data Storage Device
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video processing methods require significant host resources and data transport from data storage devices to processing destinations, as video processing is typically carried out on host processors rather than the storage devices themselves.
Innovation Solution
A chip-bound architecture within data storage devices, featuring CMOS chips coupled with NAND dies, processors, memories, and error correction engines, capable of processing video data by correlating macroblocks and determining motion vectors to generate P-frames and B-frames, thereby enabling video processing on the storage device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If video processing is carried out on host processors, then processing capability is sufficient, but data transport requirements increase and host resources are consumed
Solution Approach 1:
The data storage device acts as an intermediary between data storage and processing functions. By integrating processing capabilities directly into the storage device, the system eliminates the need for extensive data transport to external host processors, thereby reducing data transport overhead while maintaining video processing capability
Solution Approach 2:
The data storage device is designed to perform multiple functions: data storage, video processing, and frame generation. This multi-functionality allows the storage device to handle both data retention and computational tasks for video frame generation, reducing the need for separate processing systems and associated data transfers
2Power
If video processing is carried out on host processors, then processing power is available, but host memory and processor resources are consumed
Solution Approach 1:
The video processing functionality is extracted from the host processor and relocated to the data storage device. This extraction reduces the computational burden on host resources while maintaining the processing capability needed for video frame generation, effectively separating storage and processing functions
3Loss of energy
If video processing is performed locally on storage device, then data transfer is minimized, but processing capability must be integrated into storage
Solution Approach 1:
The data storage device merges storage functionality with processing capabilities by integrating CMOS chips, processors, and memories directly into the storage architecture. This combination allows the device to perform video processing locally while maintaining data storage functions, thereby minimizing data transfer overhead
Data Source
AI summary
A method and apparatus for video processing on a data storage device. A chip bound architecture includes a CMOS coupled to one or more NAND die, the CMOS including one or more processors, memories, and error correction code (ECC) engines capable of processing video data. According to certain embodiments, macroblocks are correlated between two I-frames, including motion vectors to define different locations of correlated macroblocks. A P-frame may be determined from a previous I-frame and its correlated macroblocks and motion vectors, while a B-frame may be determined from two or more adjacent I-frames with concomitant macroblocks and motion vectors, as well as P-frames associated with an adjacent I-frame.