Inline Computational Storage for Reduced Data Transfer Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing size of storage devices necessitates larger data transfers and more processing cycles, leading to latency and burden on the host processor, which could be alleviated by processing data closer to the storage device.
Innovation Solution
Incorporating a computational engine within the storage device to process data inline, allowing the host processor to issue a single command that specifies data location, operation, and result destination, enabling on-the-fly result delivery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is transferred from storage device to main memory for processing, then processing capability is improved, but data transfer time increases and host processor burden increases
Solution Approach 1:
The patent combines storage functionality and computational processing into a single integrated device. The computational storage device merges the storage medium with a computational engine that can execute processing operations directly on stored data, eliminating the need to transfer data to main memory for processing.
Solution Approach 2:
The computational storage device acts as an intermediary between the storage medium and the host processor. It receives processing commands from the host, performs the computations locally on the stored data, and returns results, thereby mediating the data flow and reducing host processor burden.
2Adaptability or versatility
If host processor executes commands to process data, then processing flexibility is improved, but host processor burden increases
Solution Approach 1:
The computational storage device performs self-service by executing processing operations autonomously on the data stored within it. The integrated computational engine can independently carry out processing commands without requiring the host processor to handle the actual computation, reducing the host's workload while maintaining processing flexibility through programmable operations.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A computational storage unit is disclosed. The computational storage unit may include a storage for a data and a controller to read the data from the storage. The computational storage unit may also include a computational engine to implement a function to process the data and generate a result. The computational storage unit may receive a command from a host processor and read the data from the storage, execute the function to process the data and generate the result, and return the result to the host processor based at least in part on the command.