Programmable Network Measurement Engine for Line-Rate Data Collection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network measurement systems are inefficient due to their inability to support flexible, scalable, and fast active and passive measurements, often requiring significant updates that degrade processing speed and do not provide byte-granularity statistics or support sophisticated measurement requirements.
Innovation Solution
A flexible programmable architecture that includes a programmable passive measurement hardware engine and an active measurement hardware engine, configured by a configuration engine, allowing for line-rate processing with minimal impact on other network components, and enabling function changes without pausing operations, by decoupling the control plane from the data plane.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If an all-in-one approach integrates all possible measurements with separate data structures and control interfaces, then measurement functionality is comprehensive, but the system becomes neither scalable nor flexible and is wasteful since not all measurement features are being used in every case
Solution Approach 1:
The system is divided into separate functional modules: a measurement function module that provides measurement capabilities, a data structure module that manages data, and a control interface module that handles control signals. This segmentation allows each module to be independently configured and used only when needed, avoiding the waste of integrating all possible measurements in every system while maintaining comprehensive functionality where required.
Solution Approach 2:
The patent creates a universal measurement framework that can adapt to different measurement requirements through configurable data structures and control interfaces. The system provides multi-functionality by allowing the same basic architecture to support various measurement types (active and passive measurements) through programmable configuration rather than requiring separate dedicated structures for each measurement type.
2Measurement precision
If all data structures are updated for each incoming packet, then measurement data is comprehensive, but processing speed degrades significantly
Solution Approach 1:
Instead of uniformly updating all data structures for every incoming packet, the system applies local quality by selectively updating only the specific data structures that are relevant to the current measurement requirements and packet type. This targeted approach maintains measurement precision for necessary data while avoiding the processing overhead of updating unnecessary data structures, thereby preserving processing speed.
Solution Approach 2:
The system performs partial action by updating only a subset of data structures based on the measurement configuration and packet characteristics, rather than exhaustively updating all possible data structures. This selective partial action maintains sufficient measurement precision for the required measurements while significantly reducing the processing burden.
3Ease of operation
If CBFV approach is used for passive measurement, then processing is simplified, but it is limited to only passive measurements and does not address active network measurement
Solution Approach 1:
The patent extends the CBFV-based passive measurement approach by integrating it with an active measurement engine within a unified programmable architecture. This universal system can perform both passive measurements (using CBFV for simplified processing) and active measurements (by injecting probe packets and analyzing responses), thereby maintaining processing simplicity where applicable while expanding versatility to support multiple measurement types through a single configurable platform.
Solution Approach 2:
The system dynamically switches between passive and active measurement modes based on configuration requirements. The programmable nature allows the measurement engine to adapt its behavior - using simplified CBFV processing for passive measurements when appropriate, and transitioning to active measurement protocols when network status information requires probe packet injection and response analysis.
4Adaptability or versatility
If reprogramming is done by integrating required network measurement function into SmartNIC or switch, then measurement capability is customized, but it is time consuming with long time to market
Solution Approach 1:
The patent implements preliminary action by pre-configuring a comprehensive set of measurement functions, data structures, and control interfaces in the programmable measurement engine before deployment. The system comes pre-loaded with multiple measurement types and configurations that can be selectively activated through software programming rather than requiring hardware reintegration, thus maintaining measurement customization capability while dramatically reducing the time to market by eliminating lengthy hardware integration and reprogramming cycles.
Data Source
AI summary
A device includes a programmable passive measurement hardware engine, a programmable active measurement hardware engine, and a configuration engine. The programmable passive measurement hardware engine is configured to collect statistical data, from data transmission at a network line rate, used for network measurement. The programmable active measurement hardware engine is configured to generate probe packets and wherein the programmable active measurement hardware engine is further configured to collect responses to the generated probe packets, wherein the collected responses are used for the network measurement. The configuration engine is configured to receive data settings and wherein the configuration engine is further configured to program the programmable passive measurement hardware engine and the programmable active measurement hardware engine with the received data settings.