Bandwidth Policing Apparatus Shared FIFO for High-Speed User Group Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing bandwidth policing methods struggle to efficiently manage a large number of user groups in IP networks, leading to increased decision processing time and accommodation costs, as they require multiple packet accumulate FIFOs and complex memory access, which hinders high-speed bandwidth control.
Innovation Solution
A bandwidth policing apparatus that accumulates packets from multiple users and user groups in a single packet accumulate FIFO, using a flow detect unit and flow group detect unit to identify and police contracted bandwidth for each flow and flow group, allowing for high-speed policing without the need to select from multiple FIFOs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple packet accumulate FIFOs are provided for each user group to enable bandwidth policing for multiple user groups, then the bandwidth control capability is improved, but the device complexity and memory access complexity increase
Solution Approach 1:
The patent merges multiple packet accumulate FIFOs into a single shared FIFO that accommodates packets from multiple user groups. Instead of providing separate FIFOs for each user group (e.g., FIFO-A for user group A, FIFO-B for user group B), the invention uses one unified FIFO where packets are accumulated and then selectively transmitted based on bandwidth policing decisions. This reduces memory access complexity while maintaining the ability to police bandwidth for multiple user groups.
Solution Approach 2:
The patent segments the bandwidth policing function into separate policing units for each user group, while sharing a common FIFO structure. Each user group has its own bandwidth policing unit that independently manages contracted bandwidth, but all packets are accumulated in a single FIFO before transmission. This segmentation of policing functions combined with merging of buffer resources achieves both versatility and reduced complexity.
2Adaptability or versatility
If multiple packet accumulate FIFOs are provided for each user group, then bandwidth policing for multiple user groups is enabled, but the decision processing time increases
Solution Approach 1:
By merging multiple FIFOs into a single shared FIFO, the patent eliminates the need to select which FIFO to access when transmitting packets. The decision processing time is reduced because instead of needing to determine which user group's FIFO to use, the system simply needs to determine whether to transmit from the single shared FIFO based on bandwidth policing results. This merging operation significantly speeds up the decision-making process.
3Adaptability or versatility
If multiple packet accumulate FIFOs are provided for each user group, then bandwidth control for each user group is achieved, but the accommodation cost increases
Solution Approach 1:
The patent merges multiple FIFO structures into a single shared FIFO, thereby reducing the total memory resources required. Instead of allocating separate FIFO buffers for each user group (which would increase accommodation cost), the system shares one FIFO buffer among all user groups, reducing the overall memory capacity requirement while maintaining the ability to provide bandwidth control for each user group through software-based policing mechanisms.
Data Source
AI summary
In a bandwidth policing apparatus which accommodates a large number of user groups, bandwidth policing for each user group is performed at high speed. When there is any extra bandwidth in bandwidth for a user group, the contracted committed information rate for each user is guaranteed invariably while using this effectively. The bandwidth policing apparatus accumulates packets of a plurality of user groups under bandwidth policing in the same packet accumulate FIFO. Also, for a packet which does not exceed the contracted bandwidth for each user, the contracted bandwidth for each user is guaranteed by deciding that the packet does not exceed the contracted bandwidth for each user group.