Simulation and deduction system based on cloud native architecture and implementation method

Abstract

The application discloses a simulation deduction system based on a cloud native architecture and an implementation method, and the system comprises a business application layer, a core service layer, a security and observability layer and an infrastructure layer; the security and observability layer uniformly governs all microservice traffic and running states; the infrastructure layer undertakes the core service layer, uniformly manages container life cycles, downwardly schedules various types of computing resource, is a core of resource scheduling and task arrangement, supports dynamic expansion and contraction of simulation tasks, rapid deployment of experiments and environment rollback; the infrastructure layer is the most bottom hardware and resource base, and provides standardized, schedulable and high-usage basic computing power support for the upper layer containers and microservices; the application improves the computing resource utilization rate from 70% to 85%, reduces 40% of infrastructure cost through containerization, improves the number of concurrent simulation tasks by 3 times, shortens the delivery cycle to 0.75, and lays a foundation for intelligent operation and maintenance, autonomous simulation and simulation as a service ecological construction.

Description

Technical Field

[0001] This invention relates to the fields of cloud-native technology and simulation technology, and more specifically, to a simulation system and implementation method based on cloud-native architecture. Background Technology

[0002] In the field of modern simulation and deduction, especially in basic simulation and deduction scenarios, simulation systems face multiple technical bottlenecks, and traditional simulation and deduction systems can no longer meet the needs of practical applications. On the one hand, traditional simulation systems mostly adopt monolithic architectures or simple distributed architectures, which suffer from poor collaboration between development and operation and maintenance, long development cycles, and low efficiency. Moreover, the operating environments of different simulation models vary greatly, easily leading to the phenomenon of "only running on my machine," which seriously affects the repeatability and scalability of simulation experiments. On the other hand, large-scale entity simulations are difficult to schedule, have unreasonable resource allocation, low infrastructure utilization, and are unable to cope with peak simulation loads. Furthermore, local failures can easily spread to the entire system, affecting the continuity and reliability of simulation and deduction.

[0003] Meanwhile, traditional simulation systems lack standardized management of environment configuration, resulting in significant issues with gradual configuration changes and an inability to guarantee the security and consistency of the simulation environment. Their service architectures are highly coupled, making module iteration difficult, and software reusability low, hindering rapid response to changing requirements. Furthermore, the lack of a robust observability system and security mechanisms makes sensitive simulation data susceptible to leakage, and anomalies difficult to locate and handle quickly. In addition, existing simulation systems largely rely on foreign operating systems and processors, with insufficient domestic adaptation, failing to meet the core requirements of independent control. They also suffer from low resource utilization, high maintenance costs, and difficulty in achieving large-scale, automated delivery.

[0004] Cloud-native technologies, with their core characteristics of elasticity, resilience, service orientation, and automation, provide an effective path to solve the above pain points. However, there is currently no integrated simulation system that deeply integrates cloud-native technologies with simulation and inference, achieving containerized deployment, intelligent scheduling, full-process automation, domestic adaptation, and high security. Therefore, there is an urgent need for a cloud-native architecture simulation system to fill the technological gap in this field, improve simulation and inference capabilities and efficiency, and meet diverse simulation and inference needs. Summary of the Invention

[0005] To address the problems existing in the background technology, this invention proposes a simulation and simulation system and its implementation method based on cloud-native architecture. Its core lies in the deep integration of cloud-native technology with simulation and simulation business. Through containerized deployment, container orchestration, microservice architecture, service governance, and full-process automation, it solves many pain points of traditional simulation and simulation systems, achieving efficient, highly reliable, highly secure, and scalable operation of simulation and simulation, while meeting the requirements for domestic adaptation.

[0006] Specifically, this invention provides a simulation and inference system based on a cloud-native architecture. The system includes a business application layer, a core service layer, a security and observability layer, and an infrastructure layer. The security and observability layer uniformly manages the traffic and operating status of all microservices. The infrastructure layer, which connects to the core service layer, uniformly manages the container lifecycle, schedules various computing resources downwards, and is the core of resource scheduling and task orchestration, supporting dynamic scaling of simulation tasks, rapid deployment of experiments, and environment rollback. The infrastructure layer is the lowest-level hardware and resource foundation, providing standardized, schedulable, and highly utilized basic computing power support for upper-layer containers and microservices.

[0007] The system adopts a hybrid cloud-native architecture, with a cloud-native microservice layer as the core support, integrating DevOps development mode and continuous delivery capabilities to achieve distributed ubiquitous computing; the system has unified IaaS capabilities and cloud services, builds a software and hardware collaborative architecture for multi-cloud governance, and achieves efficient resource scheduling and management with the application business layer as the center, assisting users in completing simulation simulations, decision support, and visualization of the situation.

[0008] Furthermore, the infrastructure layer includes a container engine and container orchestration, using Kubernetes as the core of container orchestration to manage the scheduling and lifecycle of simulation container instances. Docker container technology is used to encapsulate components such as the simulation engine, data preprocessing module, and analysis services into independent images, solidifying the dependency library versions, system configurations, and environment variables of each simulation model. It features dynamic resource allocation, automatic scaling, cluster management, unitization, and disaster recovery mechanisms, allowing for dynamic adjustment of the number of container instances based on the simulated load. This enables capacity expansion during peak periods and resource release during off-peak periods, improving infrastructure utilization and service asynchronization capabilities, and meeting the principles of cloud-native resilience.

[0009] Furthermore, the core service layer adopts cloud-native service principles to build a service system, separates modules with different lifecycles and iterates them separately, uses interface-oriented programming to achieve high internal cohesion of services, and improves software reusability by extracting common modules; based on microservice architecture, the complex simulation system is broken down into independent components such as model calculation service, data service, and visualization situation display service to achieve efficient and reliable collaboration between components.

[0010] Furthermore, the security and observability layer adopts a distributed simulation collaboration mechanism, which enables each component to dynamically perceive the system status through service registration and discovery mechanisms; configures load balancing strategies to reasonably allocate request pressure, and sets up circuit breaker isolation and rate limiting degradation mechanisms to prevent local failures from spreading to the entire simulation system, ensuring the continuity of key simulations and high-fidelity situational awareness displays.

[0011] Furthermore, the system possesses full-process automation capabilities, integrating declarative APIs, KubernetesOperator, IaC, and automated delivery tools to achieve full automation of software delivery and operation. Among these, the declarative APIs can simplify complex simulation orchestration, describe the overall expected state of simulation tasks, and are not limited to specific execution steps. Combined with the human-in-the-loop declarative instructions, the system automatically parses and executes the scheduling, enabling versioning and reusability of simulation experiment configurations, and supporting rapid adjustment of simulation parameters and scale.

[0012] Furthermore, the core service layer possesses fine-grained simulation traffic control and observability capabilities. It introduces the Istio service mesh to provide fine-grained communication control, which can route requests proportionally to different versions of the simulation model, support A / B testing and comparison of new and old algorithms, and improve the resilience of the simulation system. It protects sensitive simulation data through automatic encryption and authentication of inter-service communication, and constructs a three-in-one observable system of simulation comprehensive evaluation indicators, log tracking, and simulation records, realizing real-time collection of key indicators, distributed tracking, and rapid anomaly location.

[0013] Furthermore, the system builds a deterministic simulation foundation based on immutable infrastructure, uses runtime environment container images to define the deployment environment, ensures that the deployment instance is in a brand new state, and avoids configuration gradual change issues; it combines infrastructure to realize version control, auditing and rollback of the simulation platform environment definition to ensure the security of the simulation environment; it realizes virtualization partitioning and time-sharing reuse, improves resource utilization through multi-threaded dynamic binding; and it completes cross-platform operating system and processor adaptation.

[0014] The method for performing simulations using the aforementioned cloud-native architecture-based simulation system includes the following steps:

[0015] S1: Initialize system parameters, including configuring container images, setting container orchestration strategies, initializing microservice components, configuring environment variables and security compliance parameters, and completing infrastructure environment definition and version control;

[0016] S2: Based on Docker container technology, simulation-related components are encapsulated, model dependencies and system configurations are fixed, and deployed to a Kubernetes cluster. The scheduling and lifecycle initialization of simulation container instances are completed through a container orchestration platform.

[0017] S3: Input the expected state of the simulation task through the declarative API, and the system will automatically parse and execute the schedule, configure the microservice component collaboration mode, and enable service registration and discovery, load balancing and circuit breaker rate limiting mechanisms;

[0018] S4: Start the simulation simulation task, implement flow control and service communication encryption through the Istio service mesh, and use the observable system to collect key indicators such as simulation task Gantt chart, resource consumption, and task completion rate in real time, and record the request flow path between services.

[0019] S5: Based on the projected load, the container orchestration platform automatically adjusts the number of container instances to achieve dynamic resource scaling; the DevOps platform simultaneously completes testing and operation and maintenance monitoring to promptly detect and handle anomalies.

[0020] S6: After the simulation is completed, output the simulation results, visualized situation and comprehensive evaluation report, realize environment rollback through IaC, organize logs and tracking records, and provide data support for subsequent simulation optimization;

[0021] S7: If A / B testing or algorithm comparison is required, adjust the request routing ratio through the service mesh, repeat steps S3-S6, and complete the multi-version simulation comparison analysis.

[0022] A non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements the simulation and deduction method based on a cloud-native architecture as described above.

[0023] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the simulation and deduction method based on a cloud-native architecture as described above.

[0024] The beneficial technical effects of this invention are as follows:

[0025] (1) Architecture innovation: Adopting a hybrid cloud-native architecture, deeply integrating cloud-native technology with simulation simulation, realizing the integrated design of "containerized deployment, microservice architecture, and automated operation and maintenance", breaking through the architectural bottleneck of traditional simulation systems and improving system elasticity and resilience;

[0026] (2) Deployment and scheduling innovation: Docker containerization is used to eliminate environmental differences, and Kubernetes container orchestration is used to realize intelligent scheduling and dynamic scaling of large-scale simulation tasks, which significantly improves infrastructure utilization and solves hardware resource bottlenecks;

[0027] (3) Service governance innovation: Based on the microservice architecture, the system components are decoupled, and combined with service registration and discovery, load balancing, circuit breaking and rate limiting and other mechanisms, the system stability and continuity are guaranteed, and the degree of software reuse and iteration efficiency are improved;

[0028] (4) Automation and Observability Innovation: Integrate technologies such as IaC and declarative API to achieve full-process automation, build a three-in-one observable system, simplify simulation orchestration, achieve rapid anomaly location, and improve operation and maintenance efficiency;

[0029] (5) Localization and security innovation: Complete localization adaptation and achieve independent control of core code; improve system security and compliance through service communication encryption, environment version control and other means;

[0030] (6) Efficiency Improvement Innovation: The utilization rate of computing resources was increased from 70% to 85%, infrastructure costs were reduced by 40% through containerization, the number of concurrent simulation tasks was increased by 3 times, and the delivery cycle was shortened to 0.75, laying the foundation for intelligent operation and maintenance and autonomous simulation, and the construction of a simulation as a service ecosystem. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the overall architecture of the simulation and deduction system based on cloud-native architecture provided in Embodiment 1 of the present invention;

[0032] Figure 2 This is a schematic diagram of the containerized deployment and container orchestration process provided in Embodiment 1 of the present invention;

[0033] Figure 3 This is a schematic diagram of the microservice architecture and service governance provided in Embodiment 1 of the present invention;

[0034] Figure 4 This is a flowchart of the simulation deduction method provided in Embodiment 2 of the present invention. Detailed Implementation

[0035] Example 1

[0036] This embodiment discloses a simulation and inference system based on a cloud-native architecture provided by the present invention. The system includes a business application layer, a core service layer, a security and observability layer, and an infrastructure layer. The security and observability layer uniformly manages the traffic and operating status of all microservices. The infrastructure layer inherits from the core service layer, uniformly manages the container lifecycle, schedules various computing resources downwards, and is the core of resource scheduling and task orchestration, supporting dynamic scaling of simulation tasks, rapid deployment of experiments, and environment rollback. The infrastructure layer is the lowest-level hardware and resource foundation, providing standardized, schedulable, and highly utilized basic computing power support for upper-layer containers and microservices.

[0037] The system in this embodiment adopts a hybrid cloud-native architecture, with cloud-native microservices as its core support. It integrates DevOps development models and continuous delivery capabilities, and constructs unified computing resources in various forms, including containers, virtual machines, and functions, to achieve efficient and highly reliable distributed ubiquitous computing. The system has unified IaaS capabilities and cloud services, builds a software and hardware collaborative architecture for multi-cloud governance, and achieves efficient resource scheduling and management centered on applications. It provides one-click deployment, application-aware agile scheduling, and comprehensive monitoring and operation and maintenance capabilities. At the same time, it has data operation and security compliance assurance functions, which can assist users in completing simulation simulations, supporting decision-making, and visualizing situational awareness.

[0038] The system in this embodiment adopts a containerized deployment approach. Docker container technology encapsulates components such as the simulation engine, data preprocessing module, and analysis service into independent images, solidifying the dependency library versions, system configurations, and environment variables of each simulation model. This completely eliminates the "only runs on my machine" problem, ensuring the repeatability and elastic scalability of complex simulation experiments and meeting the cloud-native elasticity principle. Kubernetes is used as the container orchestration platform, serving as the scheduling hub for large-scale entity simulations, responsible for managing the scheduling and lifecycle of simulation container instances. In parallel simulation scenarios, the orchestration system dynamically allocates computing nodes based on resource requirements and priority strategies, achieving efficient concurrent execution of simulation tasks. It also possesses the ability to address hardware resource bottlenecks, improves service asynchronization capabilities, and flexibly utilizes cluster mode, unitization, and disaster recovery mechanisms. It features an automatic scaling mechanism that dynamically adjusts the number of container instances based on the simulation load, exhibiting peak-to-peak and off-peak resource release characteristics, significantly improving infrastructure utilization.

[0039] The system in this embodiment adopts the cloud-native service principle to build a service system, separating modules with different lifecycles and iterating them separately. It uses interface-oriented programming to achieve high cohesion within services and improves software reusability by extracting common modules. Based on a microservice architecture, the complex simulation system is broken down into independent components such as model calculation services, data services, and visualization situation display services, enabling efficient and reliable collaboration between components. It has comprehensive service governance capabilities, adopts a distributed simulation collaboration mechanism, and enables each component to dynamically perceive the system status through service registration and discovery mechanisms. It configures load balancing strategies to reasonably distribute request pressure, and sets up circuit breaker isolation and rate limiting degradation mechanisms to prevent local failures from spreading to the entire simulation system, ensuring the continuity of key simulations and high-fidelity situation display.

[0040] This embodiment's system builds a deterministic simulation foundation based on immutable infrastructure. It uses runtime environment container images to define the deployment environment, ensuring that deployment instances are always new and avoiding configuration gradient issues. Combined with Infrastructure as Code (IaC) practices, it implements version control, auditing, and rollback of the simulation platform environment definition, ensuring the security of the simulation environment. It employs core binding, NUMA-related technologies, and GPU-to-CUDA library conversion to achieve virtualization partitioning and time-sharing multiplexing, improving resource utilization through multi-threaded dynamic binding. Simultaneously, the system achieves cross-platform operating system and processor adaptation, realizing compatibility and complete domestic production on Kylin systems and Phytium chips, promoting core control of independently developed software in my country.

[0041] The system in this embodiment has full-process automation capabilities, integrating declarative APIs, Kubernetes Operator, IaC, and automated delivery tools to achieve full automation of software delivery and operation and maintenance. Among them, the declarative API can simplify complex simulation orchestration, describe the overall expected state of simulation tasks, and is no longer limited to specific execution steps. Combined with the human-in-the-loop declarative instructions, it automatically parses and executes the corresponding schedule, which greatly simplifies the orchestration of complex simulations. This enables the simulation experiment configuration to be versioned, reusable, and supports the ability to quickly adjust simulation parameters and scale.

[0042] The system in this embodiment features fine-grained simulation traffic control and observability. It introduces the Istio service mesh to provide fine-grained communication control, allowing requests to be routed proportionally to different versions of the simulation model. This supports A / B testing and comparison of new and old algorithms, enhancing the resilience of the simulation system. Sensitive simulation data is protected through automatic encryption and authentication between services. An observable system integrating simulation comprehensive evaluation metrics, log tracking, and simulation records is constructed. Real-time monitoring of key indicators such as simulation task Gantt charts, resource consumption, and task completion rates is collected, recording the flow path of requests across multiple simulation services. Distributed tracing is used to quickly locate performance bottlenecks. Standardized log formats combined with machine learning algorithms help improve the automation level of the service mesh, enabling users to monitor event anomalies in real time, grasp the overall simulation status, and quickly locate the causes of anomalies.

[0043] This embodiment of the system fully implements the DevOps philosophy, ensuring that each code commit can successfully complete unit testing and integration testing, meeting the customer's requirement for rapid response to changing needs. By using a DevOps platform to avoid software black boxes, it controls the software lifecycle from its source, achieving autonomous and transparent management of core code assets, avoiding excessive reliance in the self-development process, and realizing full-process automation, transparency, standardization, and core control over business function iteration. The system provides at least 99.999% SLA service guarantee stability, meeting the requirements of high concurrency, high throughput, and low latency, ensuring SLB performance with a CPU utilization standard deviation of less than 15% and a memory utilization coefficient of variation of less than 20%. By using a Kubernetes cluster, the work of operating and maintaining the Master node is eliminated; only the worker node online and offline procedures need to be defined. The PaaS platform enables the searching of business logs, submitting expansion tasks according to business needs, and the system automatically completes the expansion operation, reducing the business risks brought by directly operating the cluster.

[0044] Example 2

[0045] This embodiment provides a method for performing simulations using the cloud-native architecture-based simulation system described in Embodiment 1, including the following steps:

[0046] Step S1: Initialize system parameters, including configuring container images, setting container orchestration strategies, initializing microservice components, configuring environment variables and security compliance parameters, and completing the definition and version control of the Infrastructure as Code (IaC) environment;

[0047] Step S2: Based on Docker container technology, encapsulate simulation-related components, solidify model dependencies and system configurations, deploy to a Kubernetes cluster, and complete the scheduling and lifecycle initialization of simulation container instances through a container orchestration platform;

[0048] Step S3: Input the expected state of the simulation task through the declarative API. The system will automatically parse and execute the schedule, configure the microservice component collaboration mode, and enable service registration and discovery, load balancing and circuit breaker rate limiting mechanisms.

[0049] Step S4: Start the simulation simulation task, implement flow control and service communication encryption through the Istio service mesh, and use the observable system to collect key indicators such as simulation task Gantt chart, resource consumption, and task completion rate in real time, and record the request flow path between services;

[0050] Step S5: Based on the simulated load, the container orchestration platform automatically adjusts the number of container instances to achieve dynamic resource scaling; the DevOps platform simultaneously completes testing and operation and maintenance monitoring to promptly detect and handle anomalies.

[0051] Step S6: After the simulation is completed, output the simulation results, visualized situation and comprehensive evaluation report, realize environment rollback through IaC, organize logs and tracking records to provide data support for subsequent simulation optimization;

[0052] Step S7: If A / B testing or algorithm comparison is required, adjust the request routing ratio through the service mesh, repeat steps S3-S6, and complete the multi-version simulation comparison analysis.

[0053] As an example, in this embodiment, to verify the feasibility and rationality of the system of the present invention, at the technical level, the simulation components are containerized and microservices are decomposed based on a cloud-native architecture. Container orchestration and dynamic scaling are achieved through Kubernetes, service mesh management is implemented using Istio, and IaC and DevOps tools are integrated to achieve full-process automation, constructing a three-in-one observable system to ensure stable system operation. Regarding computational complexity, through containerized deployment, parallel scheduling, and resource virtualization partitioning technologies, near real-time simulation of hundreds of concurrent simulation tasks is achieved on servers with tens of cores, meeting the needs of large-scale simulation. In terms of domestic adaptation, compatibility testing between the Kylin system and Phytium chips is completed, achieving full-process domestic operation of the system and ensuring independent control of the core code.

[0054] Experimental results show that the system of this invention improves the utilization rate of computing resources from 70% to 85%, reduces infrastructure costs by 40% through containerization, increases the number of concurrent simulation tasks by 3 times, shortens the delivery cycle to 0.75, and provides at least 99.999% SLA service stability, meeting the performance requirements of high concurrency, high throughput, and low latency. The system completely solves the pain points of traditional simulation systems, such as inconsistent environments, low scheduling efficiency, easy fault propagation, and complex operation and maintenance. It can efficiently complete basic simulation inference, decision support, and visualization of situational awareness tasks, and has clear engineering application value and prospects for promotion.

[0055] In summary, the system of this invention adopts a hybrid cloud-native architecture, with cloud-native microservices as the core support, integrating DevOps development model and continuous delivery capabilities, and constructing unified computing resources including containers, virtual machines, functions, and other forms; it uses Docker containerization deployment to eliminate environmental differences, achieves intelligent scheduling and dynamic scaling of large-scale simulation tasks through Kubernetes container orchestration, ensures system stability and continuity based on microservice architecture and service governance mechanisms, ensures the security of the simulation environment by combining immutable infrastructure and IaC practices, achieves full-process automation through declarative APIs and automation tools, introduces Istio service mesh and observability system to achieve traffic control and anomaly localization, implements DevOps concepts to achieve rapid iteration and independent controllability, and completes localization adaptation. This invention solves fundamental problems of traditional simulation systems, such as inconsistent environments, inefficient scheduling, easy fault propagation, complex operation and maintenance, and insufficient localization, by deeply integrating cloud-native technology with simulation and deduction services. It fills the application gap of cloud-native architecture in the field of simulation and deduction, improves the efficiency and reliability of simulation and deduction, and lays the foundation for the construction of intelligent operation and maintenance, autonomous simulation, and simulation as a service ecosystem. It can be widely used in basic simulation and deduction as well as various complex simulation scenarios.

[0056] Example 3

[0057] This embodiment provides a non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements the simulation and deduction method based on a cloud-native architecture as described above.

[0058] Furthermore, the present invention adopts the following technical solution:

[0059] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the simulation and deduction method based on the cloud-native architecture described above.

[0060] From the above description of the embodiments, those skilled in the art will clearly understand that the facilities of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Embodiments of the present invention can be implemented using existing processors, or by dedicated processors used for this or other purposes for suitable systems, or by hardwired systems. Embodiments of the present invention also include non-transitory computer-readable storage media, comprising machine-readable media for carrying or having machine-executable instructions or data structures stored thereon; such machine-readable media can be any available medium accessible by a general-purpose or special-purpose computer or other machine with a processor. For example, such machine-readable media can include RAM, ROM, EPROM, EEPROM, CD-ROM or other optical disc storage, disk storage or other magnetic storage devices, or any other medium that can be used to carry or store the required program code in the form of machine-executable instructions or data structures and is accessible by a general-purpose or special-purpose computer or other machine with a processor. When information is transmitted or provided to a machine via a network or other communication connection (hardwired, wireless, or a combination of hardwired and wireless), that connection is also considered a machine-readable medium.

[0061] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A simulation and deduction system based on a cloud-native architecture, characterized in that, The system comprises a business application layer, a core service layer, a security and observability layer, and an infrastructure layer. The security and observability layer uniformly governs the traffic and operational status of all microservices. The infrastructure layer, which connects to the core service layer, uniformly manages the container lifecycle, schedules various computing resources downwards, and is the core of resource scheduling and task orchestration, supporting dynamic scaling of simulation tasks, rapid deployment of experiments, and environment rollback. The infrastructure layer is the lowest-level hardware and resource foundation, providing standardized, schedulable, and highly utilized basic computing power support for upper-layer containers and microservices. The system adopts a hybrid cloud-native architecture, with a cloud-native microservice layer as the core support, integrating DevOps development mode and continuous delivery capabilities to achieve distributed ubiquitous computing; the system has unified IaaS capabilities and cloud services, builds a software and hardware collaborative architecture for multi-cloud governance, and achieves efficient resource scheduling and management with the application business layer as the center, assisting users in completing simulation simulations, decision support, and visualization of the situation.

2. The simulation and deduction system based on cloud-native architecture according to claim 1, characterized in that: The infrastructure layer includes a container engine and container orchestration, using Kubernetes as the core of container orchestration to manage the scheduling and lifecycle of simulation container instances. Docker container technology is used to encapsulate components such as the simulation engine, data preprocessing module, and analysis services into independent images, and to solidify the dependency library versions, system configurations, and environment variables of each simulation model. It features dynamic resource allocation, automatic scaling, cluster management, unitization, and disaster recovery mechanisms, and can dynamically adjust the number of container instances according to the simulated load to achieve capacity expansion during peak periods and resource release during off-peak periods, improving infrastructure utilization and service asynchronization capabilities, and meeting the cloud-native resilience principle.

3. The simulation and deduction system based on cloud-native architecture according to claim 1, characterized in that: The core service layer adopts cloud-native service principles to build the service system, separates modules with different lifecycles and iterates them separately, uses interface-oriented programming to achieve high cohesion within the service, and improves software reusability by extracting common modules. Based on a microservice architecture, the complex simulation system is broken down into independent components such as model calculation services, data services, and visualization situation display services to achieve efficient and reliable collaboration between components.

4. The simulation and deduction system based on cloud-native architecture according to claim 1, characterized in that: The security and observability layer adopts a distributed simulation collaboration mechanism, which enables each component to dynamically perceive the system status through service registration and discovery mechanisms; it configures load balancing strategies to reasonably distribute request pressure, and sets up circuit breaker isolation and rate limiting degradation mechanisms to prevent local failures from spreading to the entire simulation system, ensuring the continuity of key simulations and high-fidelity situational awareness displays.

5. The simulation and deduction system based on cloud-native architecture according to claim 4, characterized in that: The system possesses full-process automation capabilities, integrating declarative APIs, Kubernetes Operator, IaC, and automated delivery tools to achieve full automation of software delivery and maintenance. Among them, the declarative API can simplify complex simulation orchestration, describe the overall expected state of simulation tasks, and is not limited to specific execution steps. Combined with the human-in-the-loop declarative instructions, it automatically parses and executes the scheduling, realizing versioning and reusability of simulation experiment configurations, and supporting rapid adjustment of simulation parameters and scale.

6. The simulation and deduction system based on cloud-native architecture according to claim 1, characterized in that: The core service layer possesses fine-grained simulation traffic control and observability capabilities. It introduces the Istio service mesh to provide fine-grained communication control, which can route requests proportionally to different versions of the simulation model. It supports A / B testing and comparison of new and old algorithms, improving the resilience of the simulation system. It protects sensitive simulation data through automatic encryption and authentication of inter-service communication, and constructs a three-in-one observable system of simulation comprehensive evaluation indicators, log tracking, and simulation records, realizing real-time collection of key indicators, distributed tracking, and rapid anomaly location.

7. The simulation and deduction system based on cloud-native architecture according to claim 1, characterized in that: The system is based on immutable infrastructure to build a deterministic simulation foundation. It uses runtime environment container images to define the deployment environment, ensuring that the deployment instance is in a brand new state and avoiding configuration gradual change issues. It combines infrastructure to realize version control, auditing and rollback of the simulation platform environment definition to ensure the security of the simulation environment. It realizes virtualization partitioning and time-sharing reuse, improves resource utilization through multi-threaded dynamic binding, and completes cross-platform operating system and processor adaptation.

8. A method for performing simulation simulations using the cloud-native architecture-based simulation system according to any one of claims 1-7, characterized in that, Includes the following steps: S1: Initialize system parameters, including configuring container images, setting container orchestration strategies, initializing microservice components, configuring environment variables and security compliance parameters, and completing infrastructure environment definition and version control; S2: Based on Docker container technology, simulation-related components are encapsulated, model dependencies and system configurations are fixed, and deployed to a Kubernetes cluster. The scheduling and lifecycle initialization of simulation container instances are completed through a container orchestration platform. S3: Input the expected state of the simulation task through the declarative API, and the system will automatically parse and execute the schedule, configure the microservice component collaboration mode, and enable service registration and discovery, load balancing and circuit breaker rate limiting mechanisms; S4: Start the simulation simulation task, implement flow control and service communication encryption through the Istio service mesh, and use the observable system to collect key indicators such as simulation task Gantt chart, resource consumption, and task completion rate in real time, and record the request flow path between services. S5: Based on the projected load, the container orchestration platform automatically adjusts the number of container instances to achieve dynamic resource scaling; the DevOps platform simultaneously completes testing and operation and maintenance monitoring to promptly detect and handle anomalies. S6: After the simulation is completed, output the simulation results, visualized situation and comprehensive evaluation report, realize environment rollback through IaC, organize logs and tracking records, and provide data support for subsequent simulation optimization; S7: If A / B testing or algorithm comparison is required, adjust the request routing ratio through the service mesh, repeat steps S3-S6, and complete the multi-version simulation comparison analysis.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the simulation and deduction method based on cloud-native architecture as described in any one of claims 1 to 8.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the simulation and deduction method based on cloud-native architecture as described in any one of claims 1 to 8.

Images