SR-MPLS Packet Tracing and Per-Hop Delay via MCD Stamping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network operators face challenges in accurately tracing the path of data packets within their networks and measuring per-hop delays and load, which complicates troubleshooting and performance optimization.
Innovation Solution
Implementing a system that computes and records midpoint compressed data (MCD) in a label stack of data packets using a path tracing indicator (PTI), allowing nodes to shift and stamp MCD at each hop, and encapsulates the data for collection, enabling path tracing and delay measurement in MPLS and SR-MPLS networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If path tracing and per-hop delay measurement are implemented in SR-MPLS networks, then network visibility and troubleshooting capability are improved, but packet processing complexity and overhead increase
Solution Approach 1:
The path tracing functionality is segmented into discrete components: a path tracing indicator (PTI) field within existing MPLS labels, separate timestamp recording mechanisms at each hop, and a collector system that aggregates data. This segmentation allows the complex path tracing function to be distributed across network elements without requiring a single complex processing system.
Solution Approach 2:
The path tracing data structures are nested within existing MPLS label stacks. The PTI field is embedded within the label header, and timestamp/compressed data structures are nested within the label stack, allowing path tracing functionality to be contained within existing packet formats without adding separate processing layers.
2Loss of information
If midpoint compressed data is recorded and shifted at each hop, then path tracing information is captured, but packet size and processing overhead increase
Solution Approach 1:
Only the essential path tracing information is extracted and recorded at each hop: timestamps, interface identifiers, and load data. The compressed data structure removes redundant information, keeping only what is necessary for path reconstruction and delay calculation, thereby minimizing the quantity of data while maintaining information completeness.
Solution Approach 2:
The data recording mechanism changes parameters dynamically based on network conditions. Compressed data structures adapt the amount of information recorded, and timestamps are synchronized based on local clock parameters. This allows the system to maintain path information completeness while optimizing packet data volume according to actual network requirements.
3Measurement precision
If per-hop delay measurement is performed, then network performance characterization is improved, but measurement overhead and processing time increase
Solution Approach 1:
Timestamps are recorded at each hop immediately upon packet arrival, and the path tracing indicator is set in advance to trigger compression and data collection. This preliminary action ensures that timing data is captured without waiting for end-to-end processing, reducing measurement processing time while maintaining per-hop delay accuracy.
Solution Approach 2:
The path tracing data collection process is optimized to skip unnecessary processing steps. Compressed data structures minimize the amount of data that needs to be processed at each hop, and the collector system rapidly aggregates and processes the compressed data, rushing through the measurement process to reduce overall processing time.
Data Source
AI summary
A network node receives a data packet. In response to receiving the data packet, the network node performs a lookup on a label stack of the data packet to determine a next hop for the data packet. The network node scans the label stack to identify a Structured Entropy Label (SEL). The SEL includes a Path Tracing Indicator (PTI). The network node computes Midpoint Compressed Data (MCD) as a result of the PTI being set to a pre-defined value. The network node records the MCD in a MCD stack of the data packet by shifting the MCD stack and stamping the MCD on top of the MCD stack. The network node transmits the data packet to the next hop with the recorded MCD stack. The network sink node encapsulates the received data packet to generate an encapsulated data packet and transmits the data packet.