Leaf Node Route Preference Markers Prevent Data Center Tromboning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In data center architectures, the inefficiency in routing inter-subnet traffic, known as tromboning, leads to increased network latency, congested network links, and higher bandwidth consumption, which negatively impacts Internet access speeds.
Innovation Solution
Implementing a method at leaf nodes to detect and reject route advertisements to spine nodes in remote data centers, allowing only local spine nodes to be learned, thereby preventing traffic from being forwarded to remote spine nodes that would otherwise return it, thus avoiding tromboning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If route advertisements to remote spine nodes are accepted, then routing flexibility is improved, but network latency increases due to tromboning
Solution Approach 1:
The patent segments the routing decision process by introducing local route preference markers that divide routes into preferred local paths and non-preferred remote paths. Leaf nodes use these markers to segment traffic flow, directing it through local spine nodes rather than allowingtromboning through remote spines, thus reducing latency while maintaining routing flexibility.
Solution Approach 2:
The patent applies preliminary action by pre-marking local spine node routes as preferred before traffic routing decisions are made. Route advertisements include preference markers that indicate whether a route passes through a local or remote spine node, allowing leaf nodes to make immediate routing decisions without real-time analysis, thereby preventing tromboning proactively.
2Adaptability or versatility
If route advertisements to remote spine nodes are accepted, then routing options increase, but network link congestion worsens
Solution Approach 1:
The patent segments network traffic into local and remote components using route preference markers. By identifying and separating traffic that should traverse local spine nodes from traffic requiring remote spine nodes, the system optimizes network link utilization and prevents congestion on links that would otherwise carry unnecessary tromboned traffic.
Solution Approach 2:
The patent applies local quality by making routing decisions based on the local/remote nature of spine nodes. Leaf nodes preferentially select routes through local spine nodes when possible, optimizing traffic flow within the local data center and reducing overall network link congestion while preserving routing options for remote access when needed.
3Adaptability or versatility
If route advertisements to remote spine nodes are accepted, then route diversity is improved, but bandwidth consumption increases
Solution Approach 1:
The patent uses preliminary action by embedding preference markers in route advertisements that pre-indicate the local or remote nature of routes. This allows receiving nodes to immediately identify and prefer local routes without consuming additional bandwidth analyzing route paths in real-time, thus reducing overall bandwidth consumption while maintaining route diversity through preserved remote routing options.
4Productivity
If route advertisements to remote spine nodes are rejected, then network efficiency is improved, but routing adaptability decreases
Solution Approach 1:
The patent applies local quality by implementing local route preference markers that enable leaf nodes to distinguish between local and remote spine node routes. This allows the system to optimize for local traffic efficiency while preserving the ability to adaptively route traffic through remote spines when necessary, thus maintaining routing adaptability without sacrificing network efficiency.
Solution Approach 2:
The patent uses preliminary action by pre-marking routes with preference indicators before routing decisions are made. This allows the network to efficiently prefer local routes while maintaining adaptability to select remote routes when required, resolving the contradiction between network efficiency and routing adaptability through advance route characterization.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The disclosed computer-implemented method for preventing tromboning in inter-subnet traffic within data center architectures may include (1) detecting, at a leaf node of a data center, a route advertisement that advertises a route to a spine node of another data center that interfaces with the data center, (2) identifying, at the leaf node, an IP identifier of the spine node included in the route advertisement, (3) determining, at the leaf node, that the route corresponds to the spine node based at least in part on the IP identifier identified in the route advertisement, and then in response to determining that the route corresponds to the spine node, (4) rejecting the route to the spine node at the leaf node such that the leaf node does not learn the route to the spine node. Various other methods, systems, and apparatuses are also disclosed.