Network Telemetry Data Optimization Using Short Interface Identifiers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional network telemetry and delay measurement solutions face issues such as significant overhead, hardware limitations, and compatibility problems due to excessive bit usage for node identifiers and timestamps, leading to increased network utilization and dropped packets.
Innovation Solution
Implementing short interface identifiers and timestamps, such as 8-bit or 12-bit IDs and timestamps, which are uniquely assigned within each router, and using flexible telemetry data carriers like IPv6 Hop-by-Hop headers or Segment Routing Headers to minimize overhead and hardware complexity, allowing for efficient path tracing and delay measurement across different network devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional 64-bit timestamps and 24-128 bit node identifiers are used for telemetry data collection, then measurement precision and path tracing accuracy are improved, but telemetry data overhead and network utilization increase significantly
Solution Approach 1:
The patent segments the 64-bit timestamp into two parts: a 32-bit upper portion (seconds since epoch) and a 32-bit lower portion (fractional seconds in nanoseconds). This segmentation allows selective use of timestamp precision based on measurement needs while reducing overall data overhead. The segmented approach enables different nodes to use different timestamp precisions appropriate to their specific measurement requirements.
Solution Approach 2:
The patent changes the parameter of timestamp bit length from fixed 64-bit to variable length (32-bit or 64-bit) based on measurement precision requirements. For delay measurements where only relative timing is needed, 32-bit timestamps suffice. For applications requiring absolute timing precision, 64-bit timestamps are used. This parameter change resolves the contradiction by adapting timestamp precision to actual measurement needs rather than using fixed high precision everywhere.
2Measurement precision
If traditional 24-128 bit node identifiers are used for path tracing, then path tracing accuracy is improved, but device complexity and hardware limitations are exceeded
Solution Approach 1:
The patent introduces a hierarchical dimension to node identification by combining a compressed node ID (8-16 bits) with a segment identifier (16-32 bits). This dimensional change allows the system to maintain path tracing accuracy through the combination of identifiers while keeping individual identifier lengths manageable for hardware processing. The hierarchical structure enables efficient routing decisions at each network node without requiring full 128-bit identifier processing at every hop.
Solution Approach 2:
The patent extracts the essential identifying information from long node identifiers by using compressed node IDs (8-16 bits) that capture the most significant routing information. The less critical portion is replaced with shorter segment identifiers. This extraction principle reduces the total identifier length from 128 bits to 24-48 bits while maintaining sufficient precision for path tracing, thereby reducing device complexity and hardware processing requirements.
3Loss of information
If comprehensive telemetry data is collected from each node, then network visibility is improved, but network utilization and packet loss increase
Solution Approach 1:
The patent applies partial action by collecting only the necessary portion of telemetry data at each node rather than comprehensive data. Specifically, nodes collect compressed identifiers (8-16 bits) and selectively collect timestamp portions (32 or 64 bits) based on measurement requirements. This partial collection approach maintains sufficient network visibility for path tracing and delay measurement while significantly reducing the total volume of telemetry data transmitted through the network, thereby reducing network utilization and packet loss.
4Measurement precision
If standard 64-bit timestamp formats are used for delay measurement, then measurement precision is improved, but telemetry data size and processing overhead increase
Solution Approach 1:
The patent changes the timestamp parameter from fixed 64-bit to variable length (32-bit or 64-bit) based on measurement precision requirements. For most delay measurements where nanosecond precision is sufficient, the patent uses 32-bit timestamps. Only when absolute timing precision is required does the system use 64-bit timestamps. This parameter change directly resolves the contradiction by adapting timestamp size to actual measurement needs, reducing telemetry data size while maintaining adequate precision for delay measurement applications.
Data Source
AI summary
Techniques for optimizing technologies related to network path tracing and network delay measurements are described herein. Some of the techniques may include using an IPv6 header option and/or segment identifier field of a segment list or a TLV of a segment routing header as a telemetry data carrier. The techniques may also include using an SRv6 micro-segment (uSID) instruction to indicate to a node of a network that the node is to perform one or more path tracing actions and encapsulating the packet and forward. Additionally, the techniques may include using short interface identifiers corresponding to node interfaces to trace a packet path through a network. Further, the techniques may include using short timestamps to determine delay measurements associated with sending a packet through a network. In various examples, the techniques described above and herein may be used with each other to optimize network path tracing and delay measurement techniques.