Storage Device Packet Descriptor Flags for Dropped-Packet Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage devices face challenges in efficiently managing dropped packets, leading to inefficiencies in communication resources, power consumption, and bandwidth usage due to the inability to accurately identify and correct error bits causing packet drops.
Innovation Solution
A storage device equipped with a log register to store packet descriptor information for dropped packets, allowing a host device to request and receive detailed error information through a packet descriptor flag and response, enabling accurate identification and correction of error bits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If packet descriptor information is stored in the storage device, then packet identification accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments packet information into two parts: packet data itself and packet descriptor information. The descriptor information is separated into a log register within the storage device and a corresponding entry in the host device's receive buffer. This segmentation allows the storage device to maintain only essential drop reason information without storing complete packet data, reducing complexity while preserving identification accuracy.
Solution Approach 2:
The patent extracts only the necessary packet descriptor information (such as drop reasons and packet identifiers) from the complete packet data and stores it separately in the log register. This extraction approach allows the storage device to provide accurate packet identification without handling the full packet payload, thereby reducing device complexity.
2Reliability
If dropped packets are retransmitted without error correction, then communication reliability is improved, but communication time increases
Solution Approach 1:
The storage device performs preliminary action by storing packet descriptor information at the time of packet drop. This pre-stored information includes the packet identifier and drop reason, which are immediately available when the host device requests error information. This preliminary preparation eliminates the need for time-consuming packet retransmission and error analysis during communication.
Solution Approach 2:
The patent implements feedback by having the storage device provide packet descriptor information to the host device through a dedicated interface. The host device receives this feedback information (packet identifier and drop reason) and uses it to correct errors without requiring full packet retransmission. This feedback mechanism significantly reduces communication time while maintaining reliability.
3Measurement precision
If complete packet data is stored for error analysis, then error identification accuracy is improved, but energy consumption increases
Solution Approach 1:
The patent extracts only the essential packet descriptor information (packet identifier and drop reason) from the complete packet data for storage and error analysis. By taking out only the necessary metadata rather than storing the entire packet payload, the system achieves accurate error identification with significantly reduced energy consumption.
Solution Approach 2:
The patent applies local quality by storing different levels of detail for different types of information. Complete packet data is stored temporarily for processing, but only the essential descriptor information is retained in the log register for long-term error analysis. This selective retention strategy reduces energy consumption by minimizing the amount of data that needs to be maintained and accessed.
Data Source
AI summary
An electronic device which including a host device and a storage device. A method of operating the electronic device includes providing, by the host device, a first message including a plurality of packets to the storage device, providing, by the host device, a first command for checking a status of the storage device to the storage device, providing, by the storage device in response to the first command, the host device with a first response including a packet descriptor flag indicating that a first packet among the plurality of packets of the first message is dropped, and providing, by the host device, the storage device with a second command requesting first packet descriptor information of the first packet based on the packet descriptor flag of the first response.