Packet-Level Redundancy Across Separate Paths for Low-Latency Delivery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed computing systems experience high transmission latency due to network congestion and hardware faults, which are costly and time-consuming to address, particularly affecting latency-sensitive applications like stock price data transmission.
Innovation Solution
Implementing packet level redundancy by duplicating packets and manipulating their headers to transmit them via different network paths, using a packet processor with a duplicator, header manipulator, and de-duplicator to ensure delivery via distinct routes, reducing the risk of packet loss and latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If packets are resent to correct data errors from hardware faults, then data reliability is improved, but transmission latency increases significantly
Solution Approach 1:
The patent applies preliminary action by duplicating packets before transmission and routing them through different paths in advance. This way, when hardware faults occur, the alternative paths are already prepared and can be used immediately without waiting for error detection and retransmission, thus reducing latency while maintaining reliability
Solution Approach 2:
The patent implements beforehand cushioning by creating redundant packet copies and distributing them across multiple network paths before any faults occur. This cushioning against hardware failures ensures that when errors are detected, the system can switch to alternative paths without significant delay, resolving the contradiction between reliability and latency
2Reliability
If network paths are made more robust to handle hardware faults, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the network path into multiple independent routes and segmenting packet transmission across these segments. Each packet copy can traverse different network segments, and the system only needs to manage simple routing decisions at each segment rather than complex coordinated recovery mechanisms, thus improving reliability without excessive complexity
Solution Approach 2:
The patent uses copying by creating duplicate packet instances that can be independently routed through different paths. Instead of managing complex dynamic routing protocols or centralized control systems, the system simply copies packets and distributes them, which is computationally lightweight and maintains low device complexity while significantly improving fault tolerance
3Reliability
If packet duplication is implemented to reduce packet loss risk, then reliability is improved, but network bandwidth consumption increases
Solution Approach 1:
The patent applies local quality by implementing packet duplication selectively based on local conditions such as network path quality, fault history, and traffic patterns. Rather than blindly duplicating all packets, the system duplicates only when and where needed, optimizing the balance between reliability improvement and bandwidth consumption
Solution Approach 2:
The patent implements partial action by duplicating packets only to the extent necessary for acceptable reliability, rather than complete duplication. The system can adjust the duplication ratio and scope based on actual network conditions, avoiding excessive bandwidth consumption while maintaining sufficient reliability to handle hardware faults
Data Source
AI summary
Techniques of packet level redundancy in distributed computing systems are disclosed herein. In one example, upon receiving an original packet to be transmitted from a source host to an application executing at a destination host, the source host generates a duplicated packet based on the received original packet. The source host can then encapsulate the original and duplicated packets with first and second outer headers having first and second header values, respectively, and transmitting the original and the duplicated packets from the source host to the destination host via a first network path and a second network path in the computer network, respectively. Then, the transmitted original and duplicated packets can be de-duplicated at the destination host before providing the de-duplicated original and duplicated packets to the application executing at the destination host.