Erasure Pool Encoding Fallback for Distributed System Stability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current distributed systems relying on direct calls to accelerator chip driver APIs for encoding and decoding operations face instability and reliability issues due to potential failures or hot plugging of the accelerator chip, leading to crashes and impacting system stability.
Innovation Solution
An encoding and decoding method based on an erasure pool that stores original data in a shared memory, allowing the process to first attempt a hard encoding and decoding operation using a hard codec plug-in, and if that fails, switch to a soft encoding and decoding operation using a soft codec plug-in, thereby ensuring continued operation even if the accelerator chip fails.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the driver API of the accelerator chip is directly called to perform encoding and decoding operation, then the encoding and decoding operation can be performed, but the stability and reliability of the distributed system will rely heavily on the accelerator chip, and if a failure or hot plugging occurs to the accelerator chip, this will lead to a failure of the encoding and decoding operation and even a crash of the distributed system
Solution Approach 1:
The patent implements a fallback mechanism by preparing a soft codec plug-in as a backup solution before the hard encoding and decoding operation is attempted. When the accelerator chip fails or is hot-plugged, the system can immediately switch to the soft codec plug-in to perform encoding and decoding operations, thus cushioning the impact of the failure and preventing system crashes.
Solution Approach 2:
The patent introduces an intermediary layer (the soft codec plug-in) between the distributed system and the accelerator chip. This intermediary allows the system to perform encoding and decoding operations through alternative means when the direct call to the accelerator chip driver API fails, thereby mediating the reliability issue.
2Productivity
If the accelerator chip is used for hard encoding and decoding operation, then the encoding and decoding operation can be performed efficiently, but when some factors cause a failure of the operation process, it will directly lead to the failure of the encoding and decoding operation and even the crash of the distributed system
Solution Approach 1:
The system prepares a soft codec plug-in in advance as a cushioning backup. When the hard encoding and decoding operation fails due to accelerator chip issues, the soft codec plug-in is activated to continue the operation, thus cushioning the failure impact and maintaining operation stability.
Solution Approach 2:
The patent changes the operational parameter from hard encoding (accelerator chip-based) to soft encoding (CPU-based) when the former fails. This parameter change allows the system to adapt to failure conditions and maintain operational continuity, resolving the reliability issue while preserving productivity.
Data Source
AI summary
The present disclosure discloses an encoding and decoding method, apparatus and device based on an erasure pool, and a computer readable storage medium. The method includes: storing, upon receiving original data that needs to be subjected to data encoding and decoding, the original data into a shared memory; sending an encoding and decoding signal to an encoding and decoding process, so that the encoding and decoding process obtains the original data from the shared memory after receiving the encoding and decoding signal, and calls a hard codec plug-in to perform a hard encoding and decoding operation; controlling, if the hard encoding and decoding operation fails, the encoding and decoding process to call a soft codec plug-in to perform a soft encoding and decoding operation; and obtaining, after the encoding and decoding process completes the encoding and decoding operation, an encoding and decoding result from the shared memory.