Packet Transfer to Shared Resource Using Queue Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network resource allocation methods, such as 'Best Effort' and 'Fair Queuing', fail to efficiently manage congestion in telecommunications networks, leading to performance degradation and unfair prioritization of applications, especially when long communications overwhelm shared resources, causing interruptions and inefficient resource use.
Innovation Solution
A method involving N processing queues with assigned durations, where packets are prioritized based on their communication flow volume, ensuring continuous transfer of packets without interrupting short communications and preventing resource monopolization by high-volume communications, using a packet classification and selection mechanism to regulate packet flow.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If 'Best Effort' packet allocation is used, then network resource sharing is simple and device complexity is low, but congestion occurs and performance deteriorates when resources are overloaded
Solution Approach 1:
The network interface divides packet transmission into multiple queues (first queue for inter-packet spacing packets, second queue for other packets). This segmentation allows different packet types to be handled differently, improving throughput without significantly increasing overall system complexity.
Solution Approach 2:
Different queues are assigned different transmission priorities and scheduling rules. The first queue receives preferential treatment when packets meet inter-packet spacing criteria, while the second queue handles remaining packets. This local differentiation optimizes throughput for specific packet types without requiring complete system redesign.
2Productivity
If 'Fair Queuing' is used to ensure equitable sharing, then resource distribution is more balanced, but all communications still suffer performance degradation during prolonged congestion
Solution Approach 1:
The system pre-establishes multiple queues with different characteristics before congestion occurs. By having the first queue ready to capture packets meeting inter-packet spacing criteria, the system proactively protects certain communications from congestion effects rather than reacting after degradation begins.
Solution Approach 2:
The system changes the scheduling parameter (queue selection) based on packet characteristics (inter-packet spacing). Packets with sufficient spacing are directed to the first queue which receives preferential treatment, while other packets go to the second queue. This dynamic parameter adjustment maintains fairness while protecting performance for certain packet types.
3Productivity
If application-based prioritization is implemented, then certain communications can be favored, but the system requires constant updating and user knowledge of applications
Solution Approach 1:
Packets self-classify into different queues based on their own characteristics (inter-packet spacing) without requiring external application identification or user input. The packet's timing property automatically determines its queue assignment, eliminating the need for complex application-level classification systems.
Solution Approach 2:
The system uses a measurable packet parameter (inter-packet spacing time) to determine queue assignment instead of requiring application identification. This parameter-based approach simplifies the system by relying on observable packet timing characteristics rather than complex application metadata or user-defined priorities.
4Productivity
If long communications are allowed to transmit freely, then resource utilization is maximized, but short communications are interrupted and resource monopolization occurs
Solution Approach 1:
The transmission system is segmented into two queues that operate with different rules. The first queue captures packets from all communications that meet the inter-packet spacing criterion, effectively segmenting traffic to prevent any single communication from monopolizing resources while ensuring continuous transmission opportunities for short communications.
Solution Approach 2:
The system implements periodic sampling of packets based on inter-packet spacing intervals. By selecting packets that arrive at regular intervals for the first queue, the system creates a periodic transmission pattern that prevents resource monopolization and ensures continuous service for multiple communications including short ones.
Data Source
Figure 1~2
Figure 3
AI summary
The invention relates to a method for transferring data packets to a shared resource (12) in a telecommunication network, that comprises the following steps: providing N processing queues (Qi, i=i to N) each associated with a respective period having a predetermined duration, where N>1; based on information indicated by a received packet, allocating said packet to one of the processing queues; adding the received packet (P13) in the queue (Qi) to which it has been allocated; selecting a processing queue from the N queues so that each non-empty queue is selected at least once during the associated period; and extracting at least one packet from the selected processing queue and transferring it in priority to the shared resource (12).