Auto-tuning RTPS Protocol for Dynamic Network Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The OMG Real-Time Publish-Subscribe Protocol (RTPS) does not provide guidelines for optimizing the sending of DATA, HEARTBEAT, and ACKNACK messages to achieve best performance and resource utilization, especially in dynamic systems where network conditions and resource availability change over time, leading to inefficiencies in bandwidth, CPU, and latency.
Innovation Solution
Implementing adaptive auto-tuning protocols to dynamically adjust the Send Window Size, Sending Rate, and batch size based on network conditions and publication rates, using software executable by a computer to optimize RTPS communications, including adjusting Send Window Size based on NACK messages, dynamically adjusting Sending Rate to avoid CPU blocking, and dynamically adjusting batch size to achieve a good throughput/latency tradeoff.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the RTPS protocol sends DATA messages reliably with retransmission mechanisms, then message delivery reliability is improved, but communication latency increases due to retransmission overhead
Solution Approach 1:
The patent implements dynamic adjustment of the send window size based on network conditions and feedback from ACKNACK messages. The send window size is not fixed but adapts continuously to changing network states, allowing the system to optimize between reliability and latency by sending appropriate numbers of DATA messages without unnecessary retransmissions
Solution Approach 2:
The patent employs feedback mechanisms where the Reader sends ACKNACK messages to the Writer indicating which samples have been received and which are missing. This feedback loop enables the Writer to adjust its transmission strategy, sending repair DATA messages only when necessary, thereby reducing unnecessary retransmission overhead and latency
2Device complexity
If the RTPS protocol uses fixed Send Window Size and Sending Rate parameters, then protocol simplicity is maintained, but bandwidth utilization and performance deteriorate under dynamic network conditions
Solution Approach 1:
The patent transforms fixed parameters into dynamic ones by implementing auto-tuning protocols that continuously adjust Send Window Size and Sending Rate based on observed network conditions. The system monitors factors such as sample delivery success rates and network traffic patterns, then adapts parameters accordingly to optimize bandwidth utilization without significantly increasing protocol complexity
Solution Approach 2:
The patent systematically changes key parameters (Send Window Size, Sending Rate, batch size) based on network conditions and performance metrics. The auto-tuning mechanism modifies these parameters dynamically, allowing the system to adapt to varying network loads and conditions, thereby improving bandwidth utilization while maintaining reasonable protocol complexity
3Productivity
If the RTPS protocol increases transmission bandwidth allocation, then throughput is improved, but CPU resource utilization increases due to additional processing overhead
Solution Approach 1:
The patent implements selective retransmission where the Writer sends repair DATA messages only for the specific samples that are missing, rather than retransmitting all samples in a window. This partial action approach reduces unnecessary CPU processing and message handling overhead while maintaining throughput by efficiently delivering only the missing critical data
4Productivity
If the RTPS protocol implements adaptive auto-tuning protocols to dynamically adjust parameters, then performance and bandwidth utilization are improved, but system complexity increases
Solution Approach 1:
The patent implements self-service mechanisms where the RTPS system automatically monitors its own performance and adjusts its parameters without external intervention. The auto-tuning protocols use built-in feedback from ACKNACK messages and network condition monitoring to self-adjust Send Window Size and Sending Rate, reducing the need for complex external configuration and management while improving performance
Data Source
AI summary
Adaptive tuning techniques are provided for data communications in an Object Management Group (OMG) Real-Time Publish Subscribe (RTPS) Protocol operable over a communication network to provide good throughput/latency tradeoff as well as efficient bandwidth utilization. With this invention, latency under high throughput conditions can be reduced several times compared with the latency obtained with traditional non-adaptive approaches.


