Queue Flushing Processor Bypassing Priority for Batch Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing queue flushing methods are inefficient when a large number of queues need to be flushed, as they prioritize queue order, leading to prolonged waiting times for queues with lower priority, which can result in delayed completion of flushing operations.
Innovation Solution
A method where a queue flushing processor scans and modifies the status of target queues without considering priority, allowing for simultaneous scanning and flushing of multiple queues, ensuring batch flushing capability and completion time improvement by using a queue information table to manage flushing status and utilizing a flushing engine for proactive queue flushing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If queue flushing is performed according to priority order, then high-priority queues are flushed first ensuring their completion, but low-priority queues wait relatively long time leading to delayed flushing completion
Solution Approach 1:
The patent segments the queue flushing process into two independent phases: a scanning phase that identifies all queues needing flushing regardless of priority, and a flushing phase that processes these queues. This segmentation allows the system to separate the identification of flushing candidates from their actual processing, enabling batch operations that reduce overall waiting time while ensuring all queues are flushed.
Solution Approach 2:
The patent implements preliminary action by first scanning and identifying all queues that need flushing before actually executing the flushing operation. The scanning phase creates a list of target queues in advance, allowing the subsequent flushing phase to process multiple queues efficiently without repeatedly checking priority conditions, thus reducing total flushing time.
2Ease of operation
If a queue with low priority waits for scheduling and dequeuing, then the queue is processed in the conventional manner, but the flushing operation cannot be completed in a timely manner when many queues need flushing
Solution Approach 1:
The patent introduces an intermediary scanning mechanism that acts as a mediator between the queue management system and the flushing operation. This scanning phase identifies all queues needing flushing and creates a target queue list, which then serves as the input for the flushing phase. This intermediary structure enables batch processing of multiple queues without disrupting conventional queue operation mechanisms.
Solution Approach 2:
The patent merges multiple queue flushing operations into a single batch processing phase. Instead of processing queues individually according to priority, the system combines the identification and flushing of multiple queues into coordinated operations, improving batch flushing capability and overall productivity.
3Reliability
If traditional queue flushing command is issued to traffic management chip, then the chip waits for queue scheduling and dequeuing before flushing, but this results in relatively low flushing speed when many queues need flushing
Solution Approach 1:
The patent applies preliminary action by implementing a scanning phase that identifies all queues needing flushing before the actual flushing operation begins. This advance identification allows the flushing phase to process multiple queues in parallel without waiting for individual queue scheduling, significantly improving flushing speed while maintaining data integrity through proper status tracking.
Solution Approach 2:
The patent segments the flushing process into distinct scanning and flushing phases, allowing the scanning phase to identify all target queues independently of the flushing execution. This segmentation enables the flushing phase to operate at higher speed by processing pre-identified queues without repeated scheduling delays, while the scanning phase ensures all necessary queues are captured.
Data Source
AI summary
A queue flushing method includes scanning, by a queue flushing processor, a flushing status of a valid queue from a queue information table to determine a target queue whose flushing status is “to be flushed” in the queue information table, where the queue information table records the flushing status of the valid queue; modifying, by the queue flushing processor, the flushing status of the target queue to “start flushing”; and flushing, by the queue flushing processor, the target queue, where the flushing status of the target queue is modified to “flushing complete” after the target queue is flushed.