Ingress Headroom Buffering for Lossless Switch Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-speed data networks face inefficiencies in processing and managing network traffic, leading to packet drops due to memory overflow, particularly in lossless packet flows, which existing technologies fail to adequately address.
Innovation Solution
Implementing ingress-based headroom buffering in switching architectures, where dedicated ingress buffers reserve memory to prevent packet drops by allocating headroom memory for lossless traffic classes and optionally supporting lossy traffic classes, thereby ensuring lossless frame delivery and conserving chip space and power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If dedicated ingress buffers are allocated for headroom buffering, then packet drops are prevented and lossless frame delivery is ensured, but chip space and power consumption increase
Solution Approach 1:
The patent segments the buffering function by implementing separate ingress buffers for different traffic classes (lossless and lossy). This allows headroom buffering to be applied selectively only to lossless traffic that requires guaranteed delivery, rather than allocating buffer space for all traffic types. The segmentation enables the system to maintain reliability for critical traffic while minimizing the total chip area required for buffering.
Solution Approach 2:
The patent applies different buffering qualities to different ingress ports and traffic classes. Ingress ports handling lossless traffic are allocated dedicated headroom buffers with guaranteed space, while ingress ports handling only lossy traffic do not require such buffers. This local differentiation ensures that buffer resources are concentrated where they provide the most value for preventing packet drops, rather than uniformly distributing buffer space across all ports.
2Reliability
If dedicated ingress buffers are allocated for headroom buffering, then packet drops are prevented and lossless frame delivery is ensured, but power consumption increases
Solution Approach 1:
The patent segments power consumption by enabling headroom buffering only for ingress ports and traffic classes that require lossless delivery. Ingress buffers are activated and powered only when needed for specific lossless traffic classes, rather than maintaining power for all buffer structures. This selective activation reduces overall power consumption while maintaining reliability for critical traffic.
Solution Approach 2:
The system implements dynamic buffer allocation where ingress buffers are activated only when lossless traffic classes are detected and deactivated when only lossy traffic is present. This self-adjusting mechanism ensures that power is consumed only when the buffering function provides actual value, rather than continuously powering all buffer structures regardless of traffic requirements.
3Area of stationary object
If main packet memory is used for buffering, then chip space is reduced, but processing efficiency and packet drop prevention capability decrease
Solution Approach 1:
The patent segments the memory architecture into dedicated ingress buffers for headroom buffering and main packet memory for general packet storage. This segmentation allows the ingress buffers to handle flow control and prevent packet drops at the ingress point, while the main packet memory handles general packet routing. The separation ensures that headroom buffering operations do not contend with main memory operations, improving overall processing efficiency.
Solution Approach 2:
The ingress buffers act as an intermediary structure between the network interface and the main packet memory. They provide a buffer zone that absorbs traffic fluctuations and prevents packet drops before packets reach the main memory. This intermediary layer improves processing efficiency by handling flow control locally at ingress, reducing the burden on the main memory system while maintaining packet drop prevention capability.
Data Source
AI summary
A network device performs ingress based headroom buffering. The network device may be configured as an output queue switch and include a main packet buffer that stores packet data according to a destination egress port. The network device may implement one or more ingress buffers associated with ingress data ports in the network device. The ingress buffers may be separate from the main packet buffer. The network device may identify a flow control condition triggered by an ingress data port, such as when an amount of data stored in the main packet buffer received through the ingress data port exceeds a fill threshold. In response, the network device may send a flow control message to a link partner to cease sending network traffic through the ingress data port. The network device may store in-flight data from the link partner in an ingress buffer instead of the main packet buffer.