Samsung electronics co., ltd. (20240345882). SYSTEMS AND METHODS FOR PROCESSING FUNCTIONS IN COMPUTATIONAL STORAGE simplified abstract
Contents
SYSTEMS AND METHODS FOR PROCESSING FUNCTIONS IN COMPUTATIONAL STORAGE
Organization Name
Inventor(s)
Jonghyeon Kim of San Jose CA (US)
Soogil Jeong of Pleasanton CA (US)
SYSTEMS AND METHODS FOR PROCESSING FUNCTIONS IN COMPUTATIONAL STORAGE - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240345882 titled 'SYSTEMS AND METHODS FOR PROCESSING FUNCTIONS IN COMPUTATIONAL STORAGE
Simplified Explanation: The patent application describes a method for performing computations near memory, where a storage device receives a request to perform a function on data, with specific operations carried out by different acceleration engines within the device.
- The method involves a processor core receiving a request to perform a function on data.
- A first processor-core acceleration engine performs the first operation on the data using custom instructions.
- A first co-processor acceleration engine then performs a second operation on the result data using custom instructions.
Key Features and Innovation:
- Utilizes acceleration engines within a storage device to perform computations near memory.
- Custom instructions are used to optimize the performance of the operations.
- Divides the processing tasks between different acceleration engines for efficient computation.
Potential Applications:
- High-performance computing tasks that require fast data processing.
- Data-intensive applications in fields like artificial intelligence, scientific research, and big data analytics.
Problems Solved:
- Improves the speed and efficiency of computations by offloading tasks to specialized acceleration engines.
- Reduces the burden on the main processor core for complex operations.
Benefits:
- Faster processing speeds for demanding computational tasks.
- Enhanced performance and efficiency in data processing.
- Optimal utilization of resources within a storage device.
Commercial Applications: The technology could be applied in industries such as cloud computing, data centers, and scientific research facilities to enhance computational capabilities and speed up data processing tasks.
Questions about the Technology: 1. How does the use of custom instructions impact the performance of the operations? 2. What are the potential limitations of dividing processing tasks between different acceleration engines?
Original Abstract Submitted
provided is a method for performing computations near memory, the method including receiving, at a processor core of a storage device, a request to perform a first function on first data, the first function including a first operation and a second operation, performing, by a first processor-core acceleration engine of the storage device, the first operation on the first data, based on first processor-core custom instructions, to generate first result data, and performing, by a first co-processor acceleration engine of the storage device, the second operation on the first result data, based on first co-processor custom instructions.
