KR-102963199-B1 - Adaptive disaster recovery method using AI-based predictive analysis and resource optimization, central control system and computer program for performing the method

KR102963199B1KR 102963199 B1KR102963199 B1KR 102963199B1KR-102963199-B1

Abstract

An adaptive disaster recovery method using artificial intelligence predictive analysis and resource optimization, along with a central control system and computer program that perform this method, collects various status data from a normal operation center and a disaster recovery center in real time, predicts potential signs of failure in advance through an artificial intelligence model, automatically executes disaster recovery procedures step-by-step according to the predicted risk level, and optimizes idle resources; thereby enabling preemptive disaster response, maximizing resource efficiency, improving reliability through automation, and operating the disaster recovery system adaptively and intelligently.

Inventors

송민영
조은석
최재민

Assignees

국방과학연구소

Dates

Publication Date: 20260508
Application Date: 20250915

Claims (10)

A method performed by a central control system that adaptively performs disaster recovery using artificial intelligence predictive analysis and resource optimization in a disaster recovery system including a Primary Center and a Disaster Recovery Center (DR Center), A step of collecting multidimensional operational data including system logs, performance indicators, application error rates, security event logs, and external threat information from the operation center; A step of predicting a failure risk level representing a quantitative failure risk score using a machine learning model including a deep learning model such as LSTM (Long Short-Term Memory) or Transformer for analyzing time-series data and an anomaly detection model such as Isolation Forest or One-Class SVM for detecting new types of anomalies that did not exist in the past, based on the collected multidimensional operational data; A step of automatically performing at least one of a predefined disaster recovery procedure based on the predicted level of risk of failure; and A step of optimizing the computing resources of the disaster recovery center based on the predicted level of risk of failure; including, Adaptive disaster recovery method using artificial intelligence predictive analytics and resource optimization.
In paragraph 1, The above prediction step is, The method comprises analyzing the collected multidimensional operational data using the machine learning model specialized in time-series data analysis or anomaly detection to calculate a quantitative failure risk score indicating the probability of a failure occurring within a predetermined time, and predicting the failure risk level. Adaptive disaster recovery method using artificial intelligence predictive analytics and resource optimization.
In paragraph 1, The above execution step is, Determining the failure risk level by comparing the predicted failure risk level with a plurality of predefined thresholds, and automatically performing at least one of a predefined disaster recovery procedure according to the failure risk level, Adaptive disaster recovery method using artificial intelligence predictive analytics and resource optimization.
In Paragraph 3, The above execution step is, If the above-mentioned risk level of failure exceeds a first threshold, an interest level control operation is performed to shorten the monitoring cycle and send a notification to the administrator. Adaptive disaster recovery method using artificial intelligence predictive analytics and resource optimization.
In Paragraph 4, The above execution step is, If the above-mentioned failure risk level exceeds a second threshold that is higher than the first threshold, the method comprises starting a virtual machine (VM) of the disaster recovery center and performing a preparatory operation to preload essential data into memory. Adaptive disaster recovery method using artificial intelligence predictive analytics and resource optimization.
In paragraph 5, The above execution step is, When the above failure risk level exceeds a third threshold that is higher than the above second threshold, a preemptive transition preparation operation is performed to accelerate final data synchronization and distribute the load by diverting a portion of user traffic to the disaster recovery center. Adaptive disaster recovery method using artificial intelligence predictive analytics and resource optimization.
In paragraph 6, The above execution step is, If the above failure risk level exceeds a fourth threshold that is higher than the third threshold, the service of the operation center is normally stopped and an automated preemptive switching operation is performed to completely switch all traffic to the disaster recovery center. Adaptive disaster recovery method using artificial intelligence predictive analytics and resource optimization.
In paragraph 1, The above optimization step is, If the predicted risk level of failure is within the normal range, the idle computing resources of the disaster recovery center are dynamically allocated to tasks for other purposes, and If the predicted risk level of the above-mentioned failure falls outside the normal range, resources allocated to other tasks are reclaimed and allocated to tasks for disaster recovery purposes. Adaptive disaster recovery method using artificial intelligence predictive analytics and resource optimization.
A computer program stored on a computer-readable storage medium for executing an adaptive disaster recovery method using artificial intelligence predictive analysis and resource optimization described in any one of paragraphs 1 through 8 on a computer.
As a central control system that performs adaptive disaster recovery using artificial intelligence predictive analytics and resource optimization in a disaster recovery system including a Primary Center and a Disaster Recovery Center (DR Center), A memory for storing one or more programs for adaptively performing disaster recovery in the above disaster recovery system; and One or more processors that perform operations to adaptively perform disaster recovery in the disaster recovery system according to one or more programs stored in the memory; Includes, The above processor is, A step of collecting multidimensional operational data including system logs, performance indicators, application error rates, security event logs, and external threat information from the operation center; A step of predicting a failure risk level representing a quantitative failure risk score using a machine learning model including a deep learning model such as LSTM (Long Short-Term Memory) or Transformer for analyzing time-series data and an anomaly detection model such as Isolation Forest or One-Class SVM for detecting new types of anomalies that did not exist in the past, based on the collected multidimensional operational data; A step of automatically performing at least one of a predefined disaster recovery procedure based on the predicted level of risk of failure; and A step of optimizing the computing resources of the disaster recovery center based on the predicted level of risk of failure; performing, Central control system.

Description

Adaptive disaster recovery method using AI-based predictive analysis and resource optimization, central control system and computer program for performing the method This document relates to disaster recovery technology for computer systems, specifically to an adaptive disaster recovery method using artificial intelligence predictive analysis and resource optimization, a central control system for executing the same, and a computer program. In particular, this document relates to an adaptive disaster recovery method using artificial intelligence predictive analysis and resource optimization, which is applied to military information systems to support the stable operation of the system even in the event of a failure, as well as a central control system and a computer program for executing the same. A typical disaster recovery (DR) system is built to enable the resumption of services at a physically separated disaster recovery center (DR Center) when a disaster (natural disasters, system failures, cyber attacks, etc.) occurs at the primary center. Traditional methods replicate data from the primary center to the disaster recovery center periodically or in real-time, and failover services based on an administrator's judgment or scripts based on predefined thresholds only after a disaster actually occurs. However, these conventional technologies have the following problems. First, reactive limitations: Since recovery begins only after a failure occurs, there are limitations in bringing the Recovery Time Objective (RTO) and Recovery Point Objective (RPO) to zero. This can lead to massive economic and social losses due to service interruptions in the event of an actual disaster. Second, resource waste: System resources (servers, storage, networks, etc.) in a disaster recovery center remain idle for most of the time, awaiting failures, incurring massive setup and operating costs. This directly impacts the increase in a company's Total Cost of Ownership. Third, complexity and human error: Service switching procedures during failures are complex, and in urgent situations requiring manual intervention by administrators, there is a possibility that errors in judgment or operational mistakes could lead to larger failures. This lowers recovery reliability and increases the burden on human resources. Therefore, there is a strong need for a new paradigm of disaster recovery technology that can predict failures in advance and respond preemptively, while simultaneously efficiently utilizing expensive disaster resources and enhancing reliability through the automation of the recovery process. FIG. 1 is a block diagram illustrating a central control system according to one embodiment of the present document. Figure 2 is a block diagram illustrating the detailed configuration of the central control system illustrated in Figure 1. FIG. 3 is a flowchart illustrating an adaptive disaster recovery method using artificial intelligence predictive analysis and resource optimization according to one embodiment of the present document. FIG. 4 is a diagram illustrating the overall architecture of a central control system from a functional perspective according to one embodiment of the present document. Figure 5 is a diagram illustrating the overall operation flow of the central control system illustrated in Figure 4. Hereinafter, embodiments of this document will be described in detail with reference to the attached drawings. The advantages and features of the embodiments of this document, and the methods for achieving them, will become clear by referring to the details described below in conjunction with the attached drawings. However, the embodiments of this document are not limited to those disclosed below but can be implemented in various different forms, and the embodiments of this document are defined only by the scope of the claims. Throughout the specification, the same reference numerals refer to the same components. Unless otherwise defined, all terms used in this specification (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which the embodiments of this document pertain. Additionally, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise. In this specification, terms such as "first," "second," etc. are used to distinguish one component from another, and the scope of rights shall not be limited by these terms. For example, the first component may be named the second component, and similarly, the second component may be named the first component. In this specification, identification symbols (e.g., a, b, c, etc.) for each step are used for convenience of explanation and do not indicate the order of the steps; the steps may occur differently from the specified order unless the context clearly indicates a specific order. That is, the steps may occur in the same order as