Data Flow Control via Symbol-Level Pause Mechanism
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data flow control methods in networks, especially in real-time networks, suffer from high overhead and sluggish responsiveness due to their coarse granularity, which can lead to symbol buffer overflow when subscribers with faster clocks transmit data to those with slower clocks, and when multiple ports receive data frames faster than they can process or forward them.
Innovation Solution
A method that allows pausing data frame transmission with a granularity of one data symbol, enabling the second subscriber to instruct the first subscriber to pause and resume data frame transmission dynamically, dividing data frames into parts as needed, and using management frames for control signals to optimize bandwidth utilization and reduce latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data frames are transmitted cyclically in real-time networks with time-controlled data flow control, then buffer overflow is prevented, but overhead increases and responsiveness becomes sluggish
Solution Approach 1:
The patent segments data frames into individual data symbols that can be transmitted and controlled independently. Each data symbol can be paused or transmitted based on real-time buffer status, replacing the coarse-grained frame-level control with fine-grained symbol-level control. This segmentation enables more responsive and efficient data flow control while preventing buffer overflow.
Solution Approach 2:
The patent implements dynamic data flow control where the transmission of data symbols can be paused or resumed based on real-time buffer status at the receiving end. This dynamic control allows the system to adapt to changing network conditions and buffer states, improving responsiveness compared to static time-controlled approaches.
2Device complexity
If data frames are transmitted with coarse granularity control, then device complexity is reduced, but responsiveness deteriorates and overhead increases
Solution Approach 1:
By segmenting data frames into individual data symbols as the smallest controllable units, the patent achieves fine-grained control without significantly increasing device complexity. The receiving end can pause transmission at the symbol level, allowing precise control with minimal overhead compared to pausing at the frame level.
Solution Approach 2:
The patent changes the granularity parameter of data flow control from frame-level to symbol-level. This parameter change enables more precise control over data transmission, reducing latency and overhead while maintaining manageable device complexity through efficient pause/resume mechanisms.
3Productivity
If data frames are transmitted from faster clock to slower clock subscribers, then data transmission speed is improved, but symbol buffer overflow occurs
Solution Approach 1:
The patent implements a feedback mechanism where the receiving subscriber (with slower clock) monitors its symbol buffer status and sends pause requests back to the transmitting subscriber (with faster clock) when the buffer approaches capacity. This feedback loop allows high-speed transmission while preventing buffer overflow through real-time flow control.
Solution Approach 2:
The receiving subscriber takes preliminary anti-action by pausing data symbol transmission before the buffer actually overflows. By monitoring buffer status and issuing pause requests in advance, the system prevents buffer overflow while maintaining high transmission speeds during periods when buffer capacity is available.
4Productivity
If multiple ports receive data frames faster than processing capacity, then data reception speed is improved, but buffer overflow occurs
Solution Approach 1:
The patent uses feedback control where the receiving subscriber monitors the fill level of its symbol buffer and sends pause requests to transmitting subscribers when the buffer approaches capacity. This allows multiple ports to receive data at high speeds while dynamically preventing buffer overflow through real-time flow control based on actual processing capacity.
Data Source
AI summary
In order to implement a method for data flow control in a network with reduced overhead, a first subscriber transmits data symbols in data frames to a second subscriber. The second subscriber forwards and/or processes the data frame. If there is a delay in forwarding and/or processing the data frame, the second subscriber instructs the first subscriber to pause the transmission of the data frames with a granularity of one data symbol. After the pause, the first subscriber continues to transmit the data frames with a granularity of one data symbol.