Network Packet Bypass for Latency Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computer network technologies experience significant latency and processing power consumption due to data processing at multiple interfaces, which becomes exacerbated with large data transmissions, despite previous techniques having only partially addressed these issues.
Innovation Solution
The method involves creating a first packet type from received data packets that overrides size limits, allowing it to bypass certain processing layers and be directly routed from a receive stack to a transmit stack, using a packet map for segmentation and checksum generation, and enabling the re-transmission of large receive offset packets as transmit segmentation offload packets, thus reducing latency and processing steps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data packets are processed through multiple network layers and interfaces, then data routing accuracy and protocol compliance are improved, but latency increases and processing power is consumed
Solution Approach 1:
The patent segments the packet processing path by identifying packets that can bypass certain network layers. Specifically, packets with specific characteristics (e.g., certain protocol types or marked packets) are routed directly from the network interface controller to the host memory without undergoing full protocol stack processing, while other packets continue through the complete processing path. This segmentation resolves the contradiction by allowing time-sensitive packets to skip unnecessary processing steps while maintaining reliability for packets that require full processing.
Solution Approach 2:
The patent introduces an intermediary mechanism in the form of a bypass path with packet classification logic. This intermediary selectively directs packets around the traditional processing path when conditions permit, acting as a mediator between the need for fast transmission and the need for proper protocol handling. The intermediary evaluates packet characteristics and determines the optimal path, resolving the contradiction between speed and reliability on a per-packet basis.
2Reliability
If data packets are processed through multiple network layers, then protocol compliance and data integrity are improved, but processing power consumption increases
Solution Approach 1:
The processing workload is segmented between two paths: a full processing path for packets requiring protocol compliance verification, and a bypass path for packets that can be transmitted with minimal processing. This segmentation reduces overall processing power consumption by avoiding redundant processing steps for packets that don't require full protocol handling, while maintaining data integrity for packets that do require such processing.
Solution Approach 2:
The system employs self-service mechanisms where packets carry information about their own processing requirements (e.g., through marking or classification). This allows the network interface controller to autonomously determine which packets can bypass processing layers without requiring intensive intervention from the host processor, thereby reducing processing power consumption while maintaining data integrity through selective processing.
3Reliability
If packet size limits are enforced at each interface, then network protocol compliance is improved, but transmission efficiency for large data sets deteriorates
Solution Approach 1:
The patent implements dynamic packet size handling where the effective packet size limit is adaptable based on the transmission path and packet characteristics. For packets taking the bypass path, larger sizes are permitted since they skip certain interface processing steps that would otherwise enforce size limits. This dynamic approach allows large data transmissions to maintain efficiency while packets requiring full protocol compliance still adhere to standard size constraints, resolving the contradiction between compliance and efficiency.
Solution Approach 2:
The system changes the parameter of packet size enforcement based on the processing path selected. Packets routed through the bypass path operate under different size parameters than those processed through the full stack, allowing large packets to be transmitted efficiently without violating protocol compliance requirements for packets that undergo complete processing. This parameter change enables flexible handling of packet sizes to optimize transmission efficiency while maintaining reliability.
Data Source
AI summary
Systems, devices and methods for improving network performance are disclosed. In particular, in one embodiment, a method of processing data packets in an electronic network is provided. The method includes receiving data packets at a receive stack of an electronic network device and creating a first packet from at least a portion of each of the set of data packets. The first packet includes an indicator that the first packet is a first packet type. Additionally, the method includes passing the first packet to one of a bridge and IP layer of the receive stack. A first packet inhibitor at the one of the bridge and IP layer is bypassed upon recognizing the first packet as a first packet type. The first packet is routed to a transmit stack of the electronic network device from the one of the bridge and IP layer.