Data Center Topology Discovery via Probe Metadata
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In data center networks, conventional control-plane approaches for topology discovery are slow to react to changes and face scalability issues, leading to hosts not having an accurate view of the network topology, which hampers operations like packet forwarding, monitoring, and debugging.
Innovation Solution
Implementing a data-plane approach where hosts use probe packets with metadata added by intermediate network devices to discover and track topology changes, allowing for faster reaction to topology changes and improved scalability without relying on a central control plane.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a control-plane approach is used for topology discovery, then centralized management and control are achieved, but the system is slow to react to topology changes and faces scalability issues
Solution Approach 1:
The patent extracts the topology discovery function from the centralized control plane and implements it in the data plane at each host. Each host independently discovers topology by sending probe packets and processing metadata from intermediate devices, eliminating the bottleneck of centralized control while maintaining accurate topology views.
Solution Approach 2:
Each host performs its own topology discovery autonomously without relying on centralized control plane assistance. Hosts self-manage their topology views by processing probe packet metadata locally, enabling fast reaction to changes and improving scalability as each host operates independently.
2Measurement precision
If probe packets with metadata are used for topology discovery, then rapid and accurate topology tracking is achieved, but network overhead increases
Solution Approach 1:
The patent merges topology discovery with existing data plane operations. Probe packets are integrated into normal network traffic flows, and metadata is added to packets that are already traversing the network, thereby achieving accurate topology discovery without creating separate dedicated discovery traffic that would increase overhead.
3Productivity
If hosts independently discover topology using probe packets, then scalability is improved and reaction speed increases, but the complexity of implementing metadata processing increases
Solution Approach 1:
The patent segments the topology discovery process into distinct components: probe packet generation at the source host, metadata addition by intermediate network devices, and metadata processing at the destination host. This segmentation allows each component to be optimized independently and simplifies implementation by distributing functionality across multiple entities rather than concentrating complexity in one place.
Data Source
AI summary
Example methods are provided a first node to perform data center network topology discovery in a data center network. One example method may comprise the first node receiving multiple probe packets that include a first probe packet and a second probe packet in response to a probing process initiated by a second node. The method may also comprise extracting, from the first probe packet, first metadata that is added by a first subset of multiple intermediate network devices and extracting, from the second probe packet, second metadata that is added by a second subset of the multiple intermediate network devices. The method may further comprise processing the first metadata and the second metadata to identify respective first forwarding path and second forwarding path from the second node to the first node.