DNS Text Record Agent Synchronization Without Third-Party Services

Solution Overview

Problem

Existing methods for synchronizing multiple agents in a computing environment incur significant processing and communication overhead, making them inefficient and costly.

Innovation Solution

An agent task scheduler utilizes a local timer and Domain Name System (DNS) text records to independently schedule tasks, updating timer values and performing tasks when due, without requiring communication with other agents or third-party services.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If third-party services (e.g., Apache ZooKeeper, etcd) are used for agent synchronization, then synchronization capability is improved, but processing overhead and communication overhead increase significantly

Engineering Contradiction:

Improvesynchronization capabilityVSAvoidprocessing overhead

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The patent extracts the synchronization coordination function from third-party services and implements it using only the existing DNS infrastructure. Agents read and update DNS text records to determine task execution timing, eliminating the need for specialized synchronization services and their associated processing overhead.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent repurposes the DNS system, traditionally used for name resolution, to also handle task scheduling and synchronization. By using DNS text records to store timing information, the system leverages an existing universal infrastructure for multiple purposes, reducing the need for additional specialized services.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Reliability

If third-party services are used for agent synchronization, then synchronization capability is improved, but communication overhead increases significantly

Engineering Contradiction:

Improvesynchronization capabilityVSAvoidcommunication overhead

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The patent extracts the synchronization coordination function from third-party services and implements it using only the existing DNS infrastructure. Agents read and update DNS text records to determine task execution timing, eliminating the need for specialized synchronization services and their associated processing overhead.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent repurposes the DNS system, traditionally used for name resolution, to also handle task scheduling and synchronization. By using DNS text records to store timing information, the system leverages an existing universal infrastructure for multiple purposes, reducing the need for additional specialized services.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Productivity

If multiple agents execute tasks independently without synchronization, then processing overhead is reduced, but task coordination accuracy deteriorates

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidtask coordination accuracy

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The patent introduces DNS text records as an intermediary between independent agents. Each agent reads the DNS record to determine if it should execute the task, and updates the record after execution. This intermediary mechanism enables accurate task coordination while agents remain independently operating, maintaining processing efficiency.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20240020156A1Independent agent synchronization using domain name system

Publication Date: 2024.01.18 SAP SE

AI summary

An agent task scheduler may include a local timer to generate a current local timer value. An agent task scheduler, independent of other agent task schedulers, may arrange to periodically perform a task scheduling process. The agent task scheduler may obtain a current Domain Name System (“DNS”) text record that includes a next timer value indicating when a task should next be executed and a prior timer value indicating when the task was last executed. If the current local timer value is greater than the next timer value in the DNS text record, the scheduler may update the next timer value in the DNS text record to the current local timer value plus a value associated with periodic performance of the task scheduling process. The scheduler may also update the prior timer value in the DNS text record to the current local timer value and perform the task.