Compound TCP Bandwidth Utilization in High-Speed Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
TCP fails to efficiently utilize network resources in high-speed and long-distance networks due to its reliance on small packet loss rates, which are unachievable, leading to inefficient bandwidth utilization and potential starvation of standard TCP flows when competing with modified protocols.
Innovation Solution
Implementing a Compound TCP system that combines multiple logical data streams into a single connection using loss-based and delay-based congestion avoidance algorithms, allowing for dynamic bandwidth management and quick recovery from rate decreases, while maintaining compatibility with legacy TCP flows.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If TCP uses a single congestion control algorithm with small packet loss rate assumption, then it can maintain simplicity and compatibility, but it fails to efficiently utilize bandwidth in high-speed and long-distance networks
Solution Approach 1:
The patent segments the single TCP connection into multiple logical data streams, each handled by different congestion control schemes (loss-based and delay-based). This segmentation allows the system to simultaneously employ multiple congestion control strategies within a single connection, improving bandwidth utilization in high-speed networks while maintaining compatibility with standard TCP protocols.
2Productivity
If modified TCP protocols use aggressive congestion control algorithms, then they can improve transmission throughput, but they cause starvation of standard TCP flows and reduce TCP friendliness
Solution Approach 1:
The patent implements dynamic congestion control by maintaining multiple logical data streams with different congestion control schemes. The system dynamically selects which scheme to use based on network conditions, allowing it to achieve high throughput when possible while falling back to standard TCP behavior for compatibility, thus maintaining TCP friendliness and fairness.
Solution Approach 2:
The patent changes the parameters of congestion control by introducing multiple state variables (cwnd for loss-based control and dwnd for delay-based control) instead of a single congestion window. This allows the system to adapt its behavior to different network conditions while maintaining compatibility with standard TCP through parameter negotiation and dynamic adjustment.
3Productivity
If TCP increases congestion window linearly for each acknowledgment, then it can fairly share bandwidth among flows, but it cannot achieve large enough window size to fill high-speed network paths
Solution Approach 1:
The patent applies preliminary action by implementing a delay-based congestion avoidance scheme that proactively increases the congestion window before packet loss occurs. By using delay information to predict available bandwidth and adjusting the window size in advance, the system can reach larger window sizes faster than traditional linear increase, reducing the time to fill high-speed network paths.
Data Source
AI summary
Systems and methods for Compound TCP are described. In one aspect, a single TCP connection is opened for communication of data over a compound stream between first and second computing devices. The compound stream includes two or more logical data streams to emulate congestion control behavior of multiple separate network (TCP) connections. Data flow over the compound stream is managed with multiple different congestion control schemes.