Samsung electronics co., ltd. (20240162917). RUNTIME RECONFIGURABLE COMPRESSION FORMAT CONVERSION WITH BIT-PLANE GRANULARITY simplified abstract

RUNTIME RECONFIGURABLE COMPRESSION FORMAT CONVERSION WITH BIT-PLANE GRANULARITY

Organization Name

samsung electronics co., ltd.

Inventor(s)

Jong Hoon Shin of San Jose CA (US)

Ardavan Pedram of Santa Clara CA (US)

Joseph Hassoun of Los Gatos CA (US)

RUNTIME RECONFIGURABLE COMPRESSION FORMAT CONVERSION WITH BIT-PLANE GRANULARITY - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240162917 titled 'RUNTIME RECONFIGURABLE COMPRESSION FORMAT CONVERSION WITH BIT-PLANE GRANULARITY

Simplified Explanation

The abstract describes a runtime bit-plane data-format optimizer for a processing element, which includes a sparsity detector and a compression-converter. The sparsity detector selects a compression-conversion format based on sparsity patterns of bit-plane data, while the compression-converter converts the data to the desired format.

• Sparsity detector selects compression-conversion format based on sparsity patterns of bit-plane data.
• Performance model used to determine the format during runtime.
• First bit-plane data stored in external memory, second data to be stored in internal memory.
• Compression-converter circuit converts first data format to second data format.

Potential Applications

The technology can be applied in image and video processing systems, where efficient data compression and conversion are crucial for performance optimization.

Problems Solved

1. Efficient data storage and processing in a processing element. 2. Real-time optimization of data formats based on sparsity patterns.

Benefits

1. Improved performance and efficiency in processing elements. 2. Reduced memory usage and bandwidth requirements. 3. Enhanced overall system speed and responsiveness.

Potential Commercial Applications

Optimized data-format conversion technology can be utilized in industries such as multimedia processing, medical imaging, and surveillance systems for enhanced data processing capabilities.

Possible Prior Art

Prior art in the field of data compression and conversion techniques, such as existing algorithms and methods for optimizing data formats in processing elements.

How does the technology handle dynamic changes in sparsity patterns during runtime?

The technology uses a performance model to adapt to changes in sparsity patterns and adjust the compression-conversion format accordingly.

What impact does the compression-conversion process have on overall system latency?

The compression-conversion process is designed to be efficient and optimized for minimal latency, ensuring smooth operation of the processing element.

Original Abstract Submitted

a runtime bit-plane data-format optimizer for a processing element includes a sparsity-detector and a compression-converter. the sparsity-detector selects a bit-plane compression-conversion format during a runtime of the processing element using a performance model that is based on a first sparsity pattern of first bit-plane data stored in a memory exterior to the processing element and a second sparsity pattern of second bit-plane data that is to be stored in a memory within the processing element. the second sparsity pattern is based on a runtime configuration of the processing element. the first bit-plane data is stored using a first bit-plane compression format and the bit-plane second data is to be stored using a second bit-plane compression format. the compression-conversion circuit converts the first bit-plane compression format of the first data to be the second bit-plane compression format of the second data.