Host Offload Data Processing to Storage Device
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud service providers face challenges in reducing system overheads of hosts, which limits the utilization of computing resources and increases costs.
Innovation Solution
A data access method and apparatus that involves a first computing device connected to a host, a second computing device, and a storage pool. The method includes receiving a second write operation command, performing data processing operations such as erasure coding or encryption on the data, and writing the processed data into the storage pool, thereby reducing system overheads of the host.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the host performs data processing operations (erasure coding, encryption, etc.), then data I/O speed can be maintained, but system overheads of the host increase
Solution Approach 1:
The patent extracts data processing operations (erasure coding, encryption, decryption) from the host and relocates them to the storage device. The storage device now performs these processing functions independently, reducing the host's system overheads while maintaining data I/O speed through direct processing at the storage level.
Solution Approach 2:
The storage device acts as an intermediary between the host and the actual data processing. Instead of the host directly performing processing operations, the storage device mediates by receiving data from the host, performing the processing operations (erasure coding, encryption), and then providing the processed data back to the host, thereby reducing host overheads.
2Productivity
If the host performs data processing operations, then data can be processed, but I/O latency increases
Solution Approach 1:
The patent extracts data processing operations from the host and performs them at the storage device level. This extraction eliminates the sequential dependency where the host must wait for processing to complete before I/O operations can proceed, thereby reducing I/O latency while maintaining processing capability.
Solution Approach 2:
The storage device performs data processing operations (erasure coding, encryption) as preliminary actions before the data is actually read or written. This preliminary processing at the storage level allows the host to initiate I/O operations earlier, reducing overall I/O latency.
3Productivity
If the host performs data processing operations, then data processing can be completed, but host resource utilization decreases
Solution Approach 1:
The patent extracts data processing functions (erasure coding, encryption, decryption) from the host and implements them at the storage device. This extraction allows the host to focus on higher-level computing tasks while the storage device handles processing operations, thereby improving host resource utilization and system adaptability.
Solution Approach 2:
The storage device serves as an intermediary that handles data processing operations, freeing the host from these tasks. This intermediary role allows the host to be more versatile in its resource utilization for actual computing workloads rather than being constrained by processing overheads.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This application discloses a data access method and apparatus, and a first computing device, and belongs to the field of computer technologies. The data access method includes: The first computing device receives a second write operation command sent by a second computing device, where the second write operation command is obtained after the second computing device preprocesses a first write operation command generated by a host, and the first write operation command is used to write to-be-written data into a storage pool; the first computing device performs a data processing operation on the to-be-written data according to the second write operation command; and the first computing device writes processed to-be-written data into the storage pool. This application reduces system overheads of the host.