Probe Packet Counter Collection in Distributed Systems

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Accurately collecting counter values from distributed counters in packet-switched systems, especially in scenarios with link aggregation groups or multi-path routing protocols, is challenging due to the complexity of monitoring and calculating performance parameters like frame loss rates across multiple links.

Innovation Solution

A method involving the use of a probe packet that is forwarded through distributed packet processors to collect and aggregate counter values, which are then delivered to a management controller to determine global counter values and calculate performance parameters, such as frame loss rates, by incorporating or storing counter values into the probe packet or local memories for later retrieval.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If centralized retrieval of counter values is implemented, then measurement accuracy is improved, but device complexity and operational overhead increase

Engineering Contradiction:

Improvecounter value accuracyVSAvoidmonitoring system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces probe packets as intermediaries that traverse through distributed packet processors to collect counter values. These probe packets act as mediators between the management controller and distributed counters, enabling accurate data collection without requiring complex centralized retrieval mechanisms. The probe packet carries identification information and accumulates counter data from multiple packet processors along its traversal path.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The distributed packet processors autonomously respond to probe packets by automatically injecting their local counter values into the probe packet. This self-service mechanism eliminates the need for complex centralized retrieval protocols, as each packet processor independently contributes its data when the probe packet visits it, simplifying the overall monitoring system architecture.

Inventive Principle:

Principle #25Self-service

2Measurement precision

If probe packets traverse all packet processors to collect counter values, then measurement completeness is improved, but loss of time increases

Engineering Contradiction:

Improveglobal counter value accuracyVSAvoidcounter collection time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The management controller pre-determines the traversal path of probe packets through the distributed packet processors before initiation. By planning the probe packet route in advance and configuring the appropriate forwarding paths, the system minimizes unnecessary traversals and ensures that probe packets visit only the relevant packet processors, thereby reducing collection time while maintaining measurement completeness.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The probe packet mechanism enables continuous monitoring by repeatedly traversing the distributed packet processors along predetermined paths. This continuous action allows the system to collect updated counter values periodically without disrupting normal packet forwarding operations, maintaining both measurement accuracy and operational efficiency over time.

Inventive Principle:

Principle #20Continuity of useful action

3Ease of operation

If counter values are stored in separate fields of probe packets, then data organization is improved, but packet size increases

Engineering Contradiction:

Improvecounter value managementVSAvoidprobe packet size

Core Design Contradiction:

Ease of operationVSVolume of moving object

Solution Approach 1:

The patent divides the probe packet into multiple fields, with each field dedicated to storing counter values from specific packet processors. This segmentation approach organizes counter data systematically, making it easier to extract and process individual counter values while maintaining a structured format that facilitates management and analysis.

Inventive Principle:

Principle #1Segmentation

4Productivity

If multiple distributed packet processors are used for link aggregation, then system capacity is improved, but measurement complexity increases

Engineering Contradiction:

Improvenetwork throughputVSAvoidcounter collection complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The probe packet mechanism serves multiple functions simultaneously: it traverses distributed packet processors, collects counter values from each processor, accumulates global statistics, and returns measurement data to the management controller. This universal approach handles link aggregation monitoring, multi-path routing measurement, and frame loss calculation through a single unified mechanism, reducing overall measurement complexity despite the distributed architecture.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS9847925B2Accurate measurement of distributed counters

Publication Date: 2017.12.19 MARVELL ASIA PTE LTD

AI summary

Aspects of the disclosure provide a method for collecting distributed counter values in a packet-switched system having multiple distributed packet processors. The method includes receiving a probe packet at a packet processor, storing a counter value corresponding to a flow processed by the packet processor for subsequent delivery to a management controller, and forwarding the probe packet to a next packet processor. The next packet processor stores a counter value of the next packet processor for subsequent delivery to the management controller.