Traffic Shaping Processor Queue Search Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In telecommunications networks, managing a large number of queues for IP traffic is inefficient due to the need to quickly identify and process non-empty queues, especially in scenarios with limited bandwidth and a high number of customers, leading to suboptimal link utilization and potential bottlenecks.
Innovation Solution
A method where the traffic shaping processor assigns data packets to specific queues and performs a scheduling operation to efficiently search and send data packets from non-empty queues, using a queue identification system to group and check queues, allowing for rapid determination of the next packet to be sent, even in systems with over 100,000 queues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a large number of queues are managed to serve many customers, then customer service coverage is improved, but queue management complexity and processing time increase
Solution Approach 1:
The patent segments the queue management process into distinct phases: identification phase (detecting non-empty queues), selection phase (choosing which queue to serve), and service phase (processing packets). This segmentation allows the system to handle large numbers of queues efficiently by focusing computational resources only on non-empty queues rather than iterating through all queues sequentially.
Solution Approach 2:
The patent performs preliminary identification of non-empty queues before the main scheduling decision. By pre-detecting which queues contain packets, the system eliminates the need to check empty queues during the main service loop, significantly reducing processing time and complexity when managing large numbers of customer queues.
2Measurement precision
If traditional queue scanning methods are used to ensure fair scheduling, then scheduling accuracy is maintained, but processing speed decreases
Solution Approach 1:
The patent extracts the identification of non-empty queues as a separate preliminary step from the main scheduling process. This extraction allows the system to maintain scheduling accuracy by systematically checking queues while improving speed by avoiding repeated checks of empty queues. The separation of identification and scheduling functions enables both precision and efficiency.
Solution Approach 2:
The patent implements a dynamic queue management approach where the scheduling process adapts based on the actual state of queues. By continuously identifying which queues are non-empty and adjusting the scheduling sequence accordingly, the system maintains fairness and accuracy while optimizing processing speed through dynamic adaptation to changing network conditions.
3Reliability
If all queues are checked sequentially to determine the next packet to send, then complete queue coverage is achieved, but time consumption increases
Solution Approach 1:
The patent performs a preliminary scan to identify all non-empty queues before executing the main scheduling loop. This preliminary action ensures complete queue coverage by detecting which queues have packets, while simultaneously reducing time consumption by eliminating the need to process empty queues during the main transmission cycle.
Solution Approach 2:
The patent enables the scheduling process to skip over empty queues by using the preliminary identification results. Instead of sequentially checking every queue and wasting time on empty ones, the system rapidly transitions between identified non-empty queues, maintaining reliability through systematic coverage while dramatically reducing time consumption.
Data Source
Figure 1~4
AI summary
The invention relates to a method for an improved traffic shaping and/or management of IP traffic in a packet processing system, especially as part of a telecommunications network, wherein the packet processing system comprises a traffic shaping processor, wherein, in order to provide for an improved traffic shaping and/or management of IP traffic, a scheduling operation is performed, by the traffic shaping processor, prior to egressing the data packets, respectively, from the packet processing system, and the method comprises the following steps: -- in a first step, non-empty queues are searched regarding a first group of queues of the plurality of different queues, comprising a natural number n of queues, this first group of queues having pieces of queue identifier information in a range from an initial queue identifier information until a final queue identifier information, -- in a second step, subsequent to the first step, the traffic shaping processor detects or decides - at a first point in time - whether one or a plurality of non-empty queues is part of the first group of queues, wherein, -- if this is the case, a specific queue identification information of a considered non-empty queue within the first group of queues is identified - at a second point in time -, the considered non-empty queue comprising at least one data packet or data packets according to a sequence of data packets within that queue, and wherein -- otherwise, a further non-empty queue is searched among the plurality of different queues, -- in a third step, subsequent to the second step, regarding the considered non-empty queue within the first group of queues, the traffic shaping processor detects or decides - at a third point in time - whether one considered data packet of the at least one data packet or data packets of the considered non-empty queue is permitted or allowed, in view of the scheduling operation of the traffic shaping processor, to be sent or egressed from the traffic shaping processor.