Network Flow Measurement Using Bit-Level Keyword Structures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network flow measurement methods face challenges in achieving high performance and commonality, as they require multiple sketch-based algorithms for different measurement tasks, leading to high processing and storage resource overheads and poor support for measuring various types of traffic, such as large flows and flow bursts, with a single sensor.
Innovation Solution
A network flow measurement method that uses a bit-level keyword measurement approach, where a network measurement device measures and reports data using a first-type data structure corresponding to a bit level of the keyword, allowing for common measurement of various types of traffic without being affected by flow distribution, and a control plane device processes this data to obtain statistical results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sketch-based algorithms are deployed for different measurement tasks, then measurement precision for various traffic types is improved, but device complexity and resource overhead increase
Solution Approach 1:
The patent applies universality by designing a single measurement sensor that can perform multiple measurement tasks simultaneously. The sensor uses a unified data structure that can capture information about large flows, small flows, and flow bursts without requiring separate specialized algorithms for each traffic type, thereby reducing device complexity while maintaining measurement precision across different traffic types
Solution Approach 2:
The patent applies segmentation by dividing the measurement data into different types (large flow data, small flow data, flow burst data) within a unified data structure. This allows the single sensor to process and measure different traffic characteristics separately while using the same underlying measurement mechanism, resolving the contradiction between measurement precision and device complexity
2Measurement precision
If multiple sketch-based algorithms are deployed for different measurement tasks, then measurement precision for various traffic types is improved, but processing and storage resource overhead increase
Solution Approach 1:
The patent reduces processing and storage resource overhead by implementing a universal measurement sensor that handles multiple measurement tasks concurrently. The unified data structure allows the sensor to extract measurements for different traffic types (large flows, small flows, bursts) from the same data stream without requiring separate processing pipelines, thereby lowering computational and storage resource consumption while maintaining measurement precision
Solution Approach 2:
The patent merges multiple measurement functions into a single sensor implementation. By combining the measurement capabilities for different traffic types into one unified data structure and processing mechanism, the system eliminates redundant processing and storage operations that would occur if separate algorithms were deployed, thus reducing overall resource overhead while preserving measurement accuracy
3Device complexity
If a single sensor is used for wide network flow measurement, then device complexity is reduced, but adaptability to measure various types of traffic deteriorates
Solution Approach 1:
The patent applies dimensionality change by organizing measurement data in a multi-dimensional unified data structure that can accommodate different traffic types. The data structure includes dimensions for large flow data, small flow data, and flow burst data, allowing a single sensor to capture diverse traffic characteristics without increasing device complexity. This dimensional organization enables the sensor to adapt to various traffic types while maintaining simplicity
4Measurement precision
If sketch-based algorithms are used for network flow measurement, then measurement precision is improved, but commonality across different measurement objectives deteriorates
Solution Approach 1:
The patent achieves commonality by designing a universal measurement sensor and unified data structure that can support multiple measurement objectives simultaneously. Instead of requiring different sketch-based algorithms for different measurement tasks, the unified structure allows the same sensor to measure large flows, small flows, and flow bursts with consistent precision, thereby improving commonality across measurement objectives while maintaining measurement accuracy
Data Source
AI summary
A network flow measurement method is applicable to a system including a network measurement device and a control plane device. The network flow measurement method includes measuring, by the network measurement device, first data, where the first data includes a first-type data structure, the first-type data structure includes first measurement information of a flow, and the first measurement information corresponds to a bit of a keyword of the flow, and sending, by the network measurement device, the first data to the control plane device.