Patent 7969880

Derivative works

Defensive disclosure: derivative variations of each claim designed to render future incremental improvements obvious or non-novel.

Active provider: Google · gemini-2.5-flash

Derivative works

Defensive disclosure: derivative variations of each claim designed to render future incremental improvements obvious or non-novel.

✓ Generated

The USPTO search results confirm the functionality of searching patents via the USPTO website. While they don't directly present the patent document itself, the context implies the patent number is valid and searchable, and the previous sections are based on the full patent text from Google Patents, which is authoritative. There are no contradictions. I will now proceed with generating the Defensive Disclosure document.

Defensive Disclosure for US Patent 7,969,880

Date of Disclosure: 2026-06-05

Subject: Derivative Works and Technical Disclosures for Network Packet Relaying with Modifiable Computational Expressions

This document outlines various technical derivations and alternative embodiments related to the core concepts disclosed in US Patent 7,969,880, "Device and method for relaying packets." The purpose of this disclosure is to establish prior art for potential incremental improvements, rendering them obvious or non-novel, and thereby limiting the scope of future patenting efforts by competitors in this field.

The central inventive concept of US 7,969,880 revolves around a network relay device or method that utilizes a modifiable computational expression (e.g., a hash function) to select physical ports or port groups for packet transmission, aiming to alleviate communication load imbalance. The following derivations expand upon this core concept across various technical axes.

Derivatives for Independent Claim 1: Network Relay Device

Claim 1: A network relay device for relaying packets, comprising: an interface module including a plurality of physical ports for connection to lines, and configured to transmit and receive packets through the lines; a computing module configured to execute a computing process with a computational expression using seed information including at least one of destination information and source information associated with a received packet; a destination search module configured to, based on the result of the computation, select a physical port for transmission of the received packet from a plurality of candidate ports among the plurality of physical ports, each of the plurality of candidate ports being able to access a destination identified by the destination information; and a modifying module configured to modify the computational expression.

Derivative 1.1: Material & Component Substitution - Optical Co-processor for Hash Computation

  • Enabling Description: A network relay device where the computing module is implemented using an optical co-processor (e.g., integrated photonics with Mach-Zehnder interferometers or ring resonators) for high-speed, low-latency hash computation. The interface module would utilize optoelectronic transceivers (e.g., silicon photonics-based transceivers operating at 400Gbps or 800Gbps). The modifying module dynamically reconfigures the optical network elements within the co-processor to alter the computational expression (e.g., by adjusting phase shifts or resonant frequencies), effectively changing the optical hash function. Seed information, consisting of packet header fields (e.g., destination IP, source IP, L4 ports), is converted to optical signals for processing.
  • Mermaid Diagram:
graph TD
    A[Interface Module - Opto-Transceivers] --> B(Packet In - Optical)
    B --> C{Optical Co-processor - Computing Module}
    C -- Seed Info (Optical) --> D[Optical Hash Function Execution]
    D -- Hash Result (Optical) --> E[Opto-Electronic Converter]
    E --> F[Destination Search Module]
    F -- Selected Optical Port --> G(Packet Out - Optical)
    G --> A
    H[Modifying Module - Optical Reconfigurator] -- Reconfigure Optical Elements --> C

Derivative 1.2: Operational Parameter Expansion - Nanoscale Quantum Hash Modulator

  • Enabling Description: A network relay device designed for nanoscale quantum communication networks, where the computing module employs a quantum circuit that generates hash values based on quantum states of seed information (e.g., entangled photons representing packet data). The interface module consists of quantum entanglement distribution ports. The modifying module dynamically adjusts quantum gate operations (e.g., by altering laser pulse sequences or trapped ion configurations) within the quantum circuit to modify the quantum computational expression. This allows for load balancing in quantum packet routing, operating at femtosecond timescales and ultra-low energy consumption due to quantum parallelism.
  • Mermaid Diagram:
graph TD
    A[Quantum Interface Module - Qubit Ports] --> B(Quantum Packet In)
    B --> C{Quantum Computing Module - Qubit Processor}
    C -- Quantum Seed State --> D[Quantum Hash Function Execution]
    D -- Quantum Hash Result --> E[Quantum Measurement Module]
    E --> F[Destination Search Module]
    F -- Selected Quantum Port --> G(Quantum Packet Out)
    G --> A
    H[Modifying Module - Quantum Gate Control] -- Adjust Quantum Gates --> C

Derivative 1.3: Cross-Domain Application - Smart Grid Energy Flow Director

  • Enabling Description: A device for directing energy flow in a dynamic smart grid. The interface module connects to various power lines (physical ports) from energy sources (e.g., solar farms, wind turbines) and to energy consumers (e.g., homes, industrial facilities). The "packets" are discrete energy demand/supply requests or actual energy flow units. The computing module calculates an optimal energy distribution pathway using a computational expression (e.g., a weighted cost function) based on seed information such as current grid load, energy prices, and source generation capacity (destination/source info). The destination search module selects a physical power line or substation group (candidate ports) to route energy. The modifying module dynamically alters the weighting factors, cost functions, or optimization algorithms within the computational expression based on real-time grid conditions, regulatory changes, or predictive analytics, to alleviate load imbalances and prevent brownouts.
  • Mermaid Diagram:
graph TD
    A[Interface Module - Power Lines/Substations] --> B(Energy Request/Flow In)
    B --> C{Computing Module - Grid Optimizer}
    C -- Grid Metrics (Seed Info) --> D[Optimization Function Execution]
    D -- Optimal Route --> E[Destination Search Module]
    E -- Selected Power Line --> F(Energy Flow Out)
    F --> A
    G[Modifying Module - Grid Policy Adjuster] -- Update Weights/Algorithms --> C

Derivative 1.4: Cross-Domain Application - Agricultural Irrigation Dispatcher

  • Enabling Description: A device for optimizing water distribution in large-scale agricultural irrigation systems. The interface module consists of multiple water valves and pump controls (physical ports) connected to irrigation lines leading to different crop zones. The "packets" are requests for specific water volumes for particular zones. The computing module uses a computational expression (e.g., a heuristic algorithm) with seed information like soil moisture levels, crop water demand, and weather forecasts (destination/source context) to determine which irrigation line to activate. The destination search module selects the appropriate water valve or group of valves (candidate ports) for water delivery. The modifying module allows for dynamic alteration of the heuristic algorithm's parameters or the selection of alternative dispatch rules based on changing environmental conditions, crop growth stages, or water availability constraints, ensuring efficient water use and preventing localized over/under-irrigation.
  • Mermaid Diagram:
graph TD
    A[Interface Module - Valves/Pump Controls] --> B(Water Request In)
    B --> C{Computing Module - Irrigation Heuristic}
    C -- Sensor Data (Seed Info) --> D[Heuristic Algorithm Execution]
    D -- Water Route Decision --> E[Destination Search Module]
    E -- Selected Valve Group --> F(Water Flow Out)
    F --> A
    G[Modifying Module - Rule Adjuster] -- Update Heuristic/Rules --> C

Derivative 1.5: Cross-Domain Application - Smart City Traffic Controller

  • Enabling Description: A device for dynamically managing traffic flow within a smart city intersection or road network. The interface module comprises multiple traffic light controllers and dynamic lane assignment systems (physical ports) connected to different road segments. The "packets" are vehicle movement requests or real-time traffic density data. The computing module employs a computational expression (e.g., a predictive traffic model) using seed information such as current traffic camera data, pedestrian crossing requests, and public transport schedules (destination/source context). The destination search module selects an optimal traffic light phase or lane configuration (candidate ports) to minimize congestion. The modifying module enables real-time adaptation of the predictive traffic model's parameters (e.g., weightings for different vehicle types, response thresholds) or the selection of entirely different traffic management strategies based on special events, emergencies, or long-term urban planning goals, thereby alleviating traffic bottlenecks.
  • Mermaid Diagram:
graph TD
    A[Interface Module - Traffic Lights/Lane Controls] --> B(Traffic Data In)
    B --> C{Computing Module - Predictive Traffic Model}
    C -- Live Data (Seed Info) --> D[Model Execution]
    D -- Optimal Signal/Lane --> E[Destination Search Module]
    E -- Selected Control Policy --> F(Traffic Flow Out)
    F --> A
    G[Modifying Module - Model Tuner] -- Update Model Params/Policy --> C

Derivative 1.6: Integration with Emerging Tech - AI-Driven Real-time Hash Optimization

  • Enabling Description: A network relay device where the modifying module incorporates an AI agent (e.g., a Reinforcement Learning algorithm) that continuously monitors network performance metrics (e.g., port utilization, latency, packet loss, queue depth) across all physical ports. The AI agent uses these metrics as feedback to dynamically adjust the parameters of the hash function (computational expression) within the computing module in real-time, without explicit human intervention. The seed information includes standard packet header fields. The AI's objective function is to minimize overall network congestion and maximize throughput, learning optimal hash configurations under varying traffic patterns and network topologies.
  • Mermaid Diagram:
graph TD
    A[Packet In] --> B{Computing Module - Dynamic Hash}
    B --> C[Destination Search Module]
    C --> D[Physical Ports - Packet Out]
    D -- Network Performance Metrics --> E[AI Modifying Module - RL Agent]
    E -- Optimal Hash Params --> B
    SubGraph Network Environment
        D
        E
    End

Derivative 1.7: Integration with Emerging Tech - IoT-Enhanced Congestion-Aware Hash

  • Enabling Description: A network relay device where the modifying module receives real-time environmental and infrastructure data from a network of distributed IoT sensors. These sensors might monitor physical load on cables, temperature of network components, or localized power consumption. This IoT data, combined with traditional network metrics, serves as additional seed information for the computing module or as input to the modifying module to select or generate a more appropriate computational expression. For instance, if an IoT sensor detects overheating in a particular cable linked to a physical port, the modifying module can instruct the computing module to deprioritize that port by altering the hash function to direct traffic away from it.
  • Mermaid Diagram:
graph TD
    A[Packet In] --> B{Computing Module - Hash Function}
    B -- Seed Info --> C[Hash Calculation]
    C --> D[Destination Search Module]
    D --> E[Physical Ports - Packet Out]
    F[IoT Sensor Network] -- Environmental/Infrastructure Data --> G[Modifying Module - IoT Data Integrator]
    G -- Hash Function Selection/Parameters --> B
    E -- Network Metrics --> G

Derivative 1.8: Integration with Emerging Tech - Blockchain-Verified Hash Configuration

  • Enabling Description: A network relay device where the modifying module interacts with a decentralized blockchain ledger. Each modification to the computational expression (e.g., hash algorithm parameters, hash function selection) is recorded as a transaction on the blockchain, providing an immutable and auditable log of all changes. Furthermore, the selection or validation of a new computational expression, particularly in multi-vendor or critical infrastructure environments, can be governed by smart contracts deployed on the blockchain, requiring consensus or pre-defined conditions to be met before a modification is applied. This ensures tamper-proof configuration and verifiable operational integrity.
  • Mermaid Diagram:
graph TD
    A[Packet In] --> B{Computing Module - Hash Function}
    B --> C[Destination Search Module]
    C --> D[Physical Ports - Packet Out]
    E[Modifying Module] --> F[Blockchain Ledger]
    F -- Configuration Updates --> E
    E -- Query/Verify --> F
    F -- Smart Contract Verification --> E

Derivative 1.9: The "Inverse" or Failure Mode - Adaptive Degradation Hash Function

  • Enabling Description: A network relay device designed with a modifying module that automatically switches the computational expression to a simplified or "safe" mode upon detection of system anomalies, component failure, or high load saturation (e.g., >90% link utilization). In this mode, the hash function might revert to a basic round-robin algorithm or prioritize a "master" physical port, ensuring graceful degradation and maintaining basic connectivity rather than complete failure. The modified computational expression minimizes computational overhead and critical resource allocation during degraded states, effectively entering a low-power or limited-functionality mode. This could involve using a reduced set of seed information (e.g., only destination MAC) for hash calculation.
  • Mermaid Diagram:
stateDiagram-v2
    state Normal_Operation {
        [*] --> Initializing
        Initializing --> Active_Hashing : System Ready
        Active_Hashing --> Active_Hashing : Stable Network
        Active_Hashing --> Degraded_Mode : Anomaly Detected / High Load
        Degraded_Mode --> Active_Hashing : Recovery / Load Reduced
    }

    state Degraded_Mode {
        [*] --> Simplified_Hashing : Failure detected
        Simplified_Hashing --> Round_Robin : Critical Failure
        Simplified_Hashing --> Master_Port_Priority : Partial Failure
    }

    Active_Hashing -- Modifying Module --> Simplified_Hashing
    Simplified_Hashing -- Modifying Module --> Active_Hashing
    Round_Robin -- Modifying Module --> Simplified_Hashing
    Master_Port_Priority -- Modifying Module --> Simplified_Hashing

Derivative 1.10: The "Inverse" or Failure Mode - Low-Power Adaptive Sampling Hash

  • Enabling Description: A network relay device featuring a modifying module that, in response to external power-saving signals or internal energy budget constraints, dynamically alters the computational expression by switching to an adaptive sampling hash technique. Instead of hashing every packet, only a statistically significant subset of packets is hashed to determine a distribution policy for a larger flow. The computing module then applies this sampled policy to subsequent packets within the flow. The frequency and granularity of the sampling (and thus hash re-computation/modification) are reduced in low-power mode, significantly lowering the power consumption of the computing module and modifying module. This means the modifying module modifies the frequency or trigger conditions of computation, effectively altering the expression's "applicability scope."
  • Mermaid Diagram:
graph TD
    A[Packet Stream In] --> B{Modifying Module - Power Mgmt}
    B -- Low Power Signal --> C{Computing Module - Adaptive Sampling Hash}
    C -- Sampled Seed Info --> D[Hash/Policy Calculation]
    D -- Policy --> E[Destination Search Module - Policy-Based]
    E --> F[Physical Ports - Packet Out]
    B -- Normal Power Signal --> G{Computing Module - Full Hash}
    G -- All Seed Info --> D
    D --> E

Derivatives for Independent Claim 9: Method for Relaying Packets

Claim 9: A method for relaying packets, comprising: executing a computing process with a computational expression using seed information including at least one of destination information and source information associated with a received packet; based on the result of the computation, selecting a port group for transmission of the received packet from among a plurality of port groups including a mutually different candidate port, each port group including one or more physical ports including a candidate port being able to access a destination identified by the destination information; and modifying the computational expression.

Derivative 9.1: Material & Component Substitution - Bio-inspired Neuro-Network Routing

  • Enabling Description: A method for packet relaying utilizing a bio-inspired neural network as the computational expression. Instead of traditional hash functions, a spiking neural network (SNN) or an artificial immune system (AIS) algorithm processes the seed information (e.g., destination and source addresses, QoS tags). The output of the SNN/AIS directly influences the probability distribution for selecting a port group. The modifying module updates the synaptic weights, neuron thresholds, or immune memory of the neural network (effectively modifying the computational expression) based on reinforcement learning signals derived from network performance feedback, mimicking biological adaptation.
  • Mermaid Diagram:
graph TD
    A(Receive Packet) --> B{Extract Seed Information}
    B --> C[Execute Bio-Inspired Network]
    C -- Spiking Patterns/Immune Response --> D{Select Port Group (Probabilistic)}
    D --> E(Transmit Packet)
    E -- Network Feedback --> F[Modify Bio-Inspired Network (Adaptive Learning)]
    F --> C

Derivative 9.2: Operational Parameter Expansion - Hypersonic Network Load Distribution

  • Enabling Description: A method for relaying packets in a hypersonic communication network, where data packets (or bursts) are transmitted at extremely high frequencies (e.g., terahertz range) and require sub-nanosecond routing decisions. The computational expression is a highly optimized, pipelined hash function implemented in ultra-low latency hardware (e.g., specialized ASIC cores) operating at clock speeds exceeding 100 GHz. The seed information is compressed and processed in parallel across multiple stages of the hash function. The modifying module facilitates "on-the-fly" modification of the hash function's lookup tables or bit-manipulation parameters without interrupting the packet flow, achieving dynamic load balancing at unprecedented speeds.
  • Mermaid Diagram:
sequenceDiagram
    participant P as Packet In
    participant CM as Computing Module (Pipelined Hash ASIC)
    participant MSM as Modifying Module (High-Freq Control)
    participant DSM as Destination Search Module
    participant PG as Port Group
    participant PO as Packet Out

    P->>CM: Packet/Seed Info (Terahertz)
    loop Sub-nanosecond processing
        CM->>CM: Execute Pipelined Hash (Computational Expression)
        CM->>DSM: Hash Result
        DSM->>PG: Select Port Group
        PG->>PO: Transmit Packet
    end
    MSM->>CM: Modify Hash Params (On-the-fly, high-freq)
    Note over MSM,CM: Modifying computational expression without flow interruption

Derivative 9.3: Cross-Domain Application - Inter-planetary Data Routing

  • Enabling Description: A method for routing data packets across an interplanetary communication network, where port groups represent different communication relays (e.g., Mars orbiters, deep-space probes) or ground stations on Earth, each with vastly different latency and bandwidth characteristics. The computational expression is a multi-objective optimization function that considers factors like signal propagation delay, available power, antenna pointing, and data priority, using seed information such as celestial body positions, data origin/destination, and mission criticality. The modifying module periodically updates the weights and constraints within this optimization function, or selects entirely different routing strategies, in response to changing orbital mechanics, solar weather events, or mission phase transitions, to ensure reliable data delivery over immense distances.
  • Mermaid Diagram:
graph TD
    A(Receive Inter-planetary Packet) --> B{Extract Celestial Seed Info}
    B --> C[Execute Multi-Objective Optimization (Computational Expression)]
    C -- Optimal Route Scores --> D{Select Port Group (Inter-planetary Relay)}
    D --> E(Transmit Data)
    E -- Orbital Mechanics/Space Weather --> F[Modify Optimization Weights/Strategy]
    F --> C

Derivative 9.4: Cross-Domain Application - Pharmaceutical Compound Distribution

  • Enabling Description: A method for optimizing the distribution of pharmaceutical compounds within an automated drug synthesis and delivery system. Port groups correspond to different synthesis reactors, purification units, or storage facilities, each with varying capacities, temperatures, and contamination risks. The computational expression is a rule-based expert system or a predictive model that uses seed information such as compound purity requirements (destination info), available reagent stock (source info), environmental conditions, and production schedules. The modifying module dynamically updates the rules, confidence factors, or model parameters within the computational expression based on real-time sensor feedback from the synthesis process, quality control results, or changes in regulatory compliance, to efficiently route compounds and minimize waste or delays.
  • Mermaid Diagram:
flowchart TD
    A[Raw Material In] --> B(Receive Compound Request)
    B --> C{Extract Seed Info (Purity, Reagents, Schedule)}
    C --> D[Execute Rule-Based Expert System (Computational Expression)]
    D -- Optimal Path Score --> E{Select Port Group (Synthesis/Purification/Storage)}
    E --> F(Compound Processing/Output)
    F -- QC Feedback/Env. Data --> G[Modify Rules/Model Parameters]
    G --> D

Derivative 9.5: Cross-Domain Application - Digital Twin Resource Allocation

  • Enabling Description: A method for allocating virtual resources (e.g., compute, memory, storage) to workloads within a digital twin simulation environment. Port groups represent clusters of virtual machines, containers, or serverless functions, each with unique performance characteristics and cost profiles. The computational expression is a dynamic programming algorithm or a heuristic scheduler that utilizes seed information such as workload demands, desired QoS levels (destination info), available physical hardware capacity (source info), and current resource utilization. The modifying module continuously updates the cost functions, weighting parameters, or the entire scheduling algorithm of the computational expression based on real-time performance of the digital twin, changes in user priorities, or detection of resource contention, to ensure optimal resource allocation and avoid simulation bottlenecks.
  • Mermaid Diagram:
graph TD
    A(Receive Digital Twin Workload) --> B{Extract Resource Seed Info}
    B --> C[Execute Dynamic Programming/Heuristic (Computational Expression)]
    C -- Optimal Allocation Score --> D{Select Port Group (Virtual Resource Cluster)}
    D --> E(Deploy Workload)
    E -- Real-time DT Performance --> F[Modify DP/Heuristic Params/Algorithm]
    F --> C

Derivative 9.6: Integration with Emerging Tech - Federated Learning for Decentralized Hash Tuning

  • Enabling Description: A method where multiple network relay devices (each implementing independent claim 9) collaboratively, but privately, tune their computational expressions using federated learning. Each device's modifying module locally trains a model to optimize its hash function based on its own traffic patterns and performance metrics. Instead of sharing raw data, these devices periodically share aggregated model updates (gradients or weights) with a central orchestrator or a blockchain. The orchestrator then aggregates these updates into a global model, which is sent back to each modifying module. This global model then informs or directly modifies the computational expression (e.g., hash function parameters or selection logic) on each device, ensuring network-wide load balancing while preserving data privacy.
  • Mermaid Diagram:
sequenceDiagram
    participant D1 as Device 1
    participant D2 as Device 2
    participant DN as Device N
    participant O as Orchestrator/Blockchain

    D1->>D1: Local Traffic Analysis
    D1->>D1: Local Hash Optimization (Computational Expression)
    D1->>O: Share Model Updates
    D2->>D2: Local Traffic Analysis
    D2->>D2: Local Hash Optimization (Computational Expression)
    D2->>O: Share Model Updates
    DN->>DN: Local Traffic Analysis
    DN->>DN: Local Hash Optimization (Computational Expression)
    DN->>O: Share Model Updates
    O->>O: Aggregate Global Model
    O->>D1: Send Global Model Update
    O->>D2: Send Global Model Update
    O->>DN: Send Global Model Update
    D1->>D1: Modify Computational Expression
    D2->>D2: Modify Computational Expression
    DN->>DN: Modify Computational Expression

Derivative 9.7: Integration with Emerging Tech - Digital Twin Predictive Hash Modification

  • Enabling Description: A method for relaying packets where a digital twin of the physical network environment continuously simulates potential traffic scenarios and predicts the efficacy of various computational expressions (hash functions) before deploying them. The modifying module for the physical network device dynamically receives recommendations for computational expression changes from the digital twin. This prediction is based on the digital twin running seed information through various hash configurations and evaluating simulated performance metrics. When the modifying module applies a modification, the digital twin is immediately updated with the new configuration, creating a closed-loop predictive optimization system.
  • Mermaid Diagram:
graph TD
    A(Physical Packet In) --> B{Computing Process - Active Hash}
    B --> C[Select Port Group]
    C --> D(Physical Packet Out)
    D -- Real-time Metrics --> E[Digital Twin (Network Model)]
    E -- Simulated Traffic --> F[DT Hash Evaluator (Test Computational Expressions)]
    F -- Predicted Best Hash --> G[Modifying Module]
    G -- Modify Active Hash --> B
    B -- Current Hash Config --> E

Derivative 9.8: The "Inverse" or Failure Mode - Contingency Hash Chain Activation

  • Enabling Description: A method for packet relaying that includes a predefined "contingency hash chain" of alternative computational expressions. Upon detection of a severe network anomaly (e.g., complete failure of a port group, critical link degradation, or a security breach), the modifying module is configured to automatically activate the next hash function in the contingency chain. Each subsequent hash in the chain is designed for increasing levels of network degradation, progressively simplifying distribution logic or prioritizing specific critical services, thus enabling a controlled and predictable failover mechanism. The seed information used for hash calculation might also be dynamically reduced in complexity to ensure rapid computation in stressed conditions.
  • Mermaid Diagram:
stateDiagram-v2
    state Normal_Operation {
        [*] --> Active_Hash_1
        Active_Hash_1 --> Active_Hash_1 : Stable Network
        Active_Hash_1 --> Modifying_Module_A : Anomaly Detected
    }

    state Contingency_Mode {
        Modifying_Module_A --> Active_Hash_2 : Activate Contingency Hash 2
        Active_Hash_2 --> Active_Hash_2 : Degraded Operation
        Active_Hash_2 --> Modifying_Module_B : Further Anomaly
        Modifying_Module_B --> Active_Hash_3 : Activate Contingency Hash 3
        Active_Hash_3 --> Active_Hash_3 : Critical Operation
        Active_Hash_3 --> Modifying_Module_C : System Recovery
        Modifying_Module_C --> Active_Hash_1 : Restore Normal Operation
    }

    Modifying_Module_A --> Modifying_Module_B : Fallback Logic
    Modifying_Module_B --> Modifying_Module_C : Fallback Logic

Combination Prior Art Scenarios

These scenarios combine the teachings of US Patent 7,969,880 with existing open-source standards, demonstrating how the core inventive concept would be obvious when integrated into widely known networking frameworks.

  1. Modifiable Hash Load Balancing in Open vSwitch (OvS):

    • Description: The modifying module and the concept of altering the computational expression for load balancing (as described in US 7,969,880) are integrated into Open vSwitch (OvS), an open-source virtual switch widely used in virtualized network environments and cloud platforms. OvS already supports various load balancing algorithms (e.g., LACP bonding modes that use source MAC, destination MAC, source IP, destination IP, TCP/UDP ports as seed information for hashing across aggregated links/ports). The combination involves implementing the modifying module as a dynamically loadable OvS module or an OpenFlow controller application. This module would expose an API for administrators or an intelligent agent to select from a set of pre-defined hash functions (computational expressions) or to inject custom hash logic (by modifying parameters) for its LAGs or ECMP (Equal-Cost Multi-Path) routes, adapting to virtual network traffic patterns to alleviate congestion in a manner similar to the patent.
    • Open-Source Standard: Open vSwitch (OvS), OpenFlow Protocol.

  2. Dynamic Hash Tuning for Linux Kernel Networking (Bonding/ECMP):

    • Description: The methodology of dynamically modifying the computational expression for selecting physical ports or port groups is applied to the Linux kernel's networking stack, specifically for its network bonding (link aggregation) and Equal-Cost Multi-Path (ECMP) routing features. The existing Linux kernel allows configuration of load balancing modes (e.g., balance-xor, 802.3ad) which use various fields as seed information for hashing. The modifying module would be implemented as a kernel module or an extended iproute2 utility, allowing system administrators or an automated script to change the hash policy (computational expression) of a bonding interface or ECMP route at runtime without requiring a kernel recompilation or reboot. This could involve changing the XOR mask, adding/removing fields from the hash input, or switching between different hash algorithms (e.g., CRC32, Jenkins hash) based on observed traffic distribution.
    • Open-Source Standard: Linux Kernel Networking (netfilter, iproute2), IEEE 802.3ad Link Aggregation (LACP).

  3. AI-driven Adaptive Hashing in Data Plane Development Kit (DPDK):

    • Description: The method of executing a computing process with a computational expression using seed information and modifying that expression is incorporated into an application built using the Data Plane Development Kit (DPDK). DPDK is an open-source set of libraries for fast packet processing on commodity hardware. DPDK-based applications (e.g., software routers, firewalls) often implement their own highly optimized hash functions for flow classification or load balancing across multiple CPU cores or network interfaces (port groups). The modifying module is realized as an intelligent DPDK application component (e.g., a software thread) that monitors core utilization, packet drops, and latency. An integrated AI/ML model within this component, using these performance metrics as feedback, would then dynamically adjust the parameters (e.g., seed values, mask, shift operations) or select alternative hash functions within the DPDK's rte_hash library or custom hash implementations to optimize packet distribution and prevent bottlenecks on specific CPU cores or NIC queues.
    • Open-Source Standard: Data Plane Development Kit (DPDK).

Generated 6/5/2026, 6:02:10 PM