Distributed simulation platform elastic scheduling method and system based on cloud computing

Abstract

The invention discloses a distributed simulation platform elastic scheduling method and system based on cloud computing, and relates to the technical field of computer systems. According to the distributed simulation platform elastic scheduling method and system based on cloud computing, a cross-domain dynamic dependency processing module, a real-time task top-speed scheduling module, a mixed cloud resource adaptation module and an elastic scheduling core engine are included; and the cross-domain dynamic dependency processing module is used for realizing dynamic processing of a dependency relationship in multi-domain collaborative simulation, cross-domain collaborative allocation of resources and breakpoint self-healing of a task chain. According to the distributed simulation platform elastic scheduling method and system based on cloud computing, dynamic analysis of a multi-domain dependency relationship, cross-domain collaborative allocation of resources and breakpoint self-healing of a task chain are achieved through the cross-domain dynamic dependency processing module, the problem that a cross-domain dependency processing mechanism is rigid in the prior art is solved, and the distributed simulation platform elastic scheduling method and system based on cloud computing are high in practicability. The cross-domain data synchronization delay is reduced, and the resource utilization rate and the stability of the task chain are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer systems, and in particular to a distributed simulation platform elastic scheduling method and system based on cloud computing. Background Art

[0002] With the rapid development of fields such as digital twin cities, cloud computing-based distributed simulation platforms are playing an important role in refined urban management and intelligent decision-making. These platforms need to implement multi-domain collaborative simulation and build virtual images that are synchronized with physical entities to support city-level dynamic simulation. However, existing technologies have significant shortcomings in cross-domain dependency processing and real-time task scheduling. Cross-domain dependency processing often relies on manually preset static rules, which cannot adapt to dynamically changing scenarios and lead to data synchronization delays. Resource allocation uses a domain-independent resource allocation model that cannot be dynamically adjusted, resulting in resource waste. The task chain formed by multi-domain collaboration lacks a breakpoint self-healing mechanism. Delays in any link will trigger a chain reaction, affecting decision-making accuracy. Existing systems often use resource load threshold triggering to respond to real-time task scheduling. This results in excessive delays from when the load exceeds the threshold to when resources are ready, making it impossible to meet the timeliness requirements of real-time tasks. The resource pool lacks a preheating mechanism, making it difficult to cope with the sudden and short-term peak demand for resources from real-time tasks. In multi-tasking concurrent scenarios, the lack of a refined preemption mechanism for high-priority tasks leads to delays in resource acquisition for high-priority tasks. Furthermore, existing preemption strategies can easily cause the status of interrupted tasks to be lost, further exacerbating resource waste and delays. Furthermore, in a hybrid cloud environment, cross-domain resource scheduling and security isolation conflict, and encryption performance loss in sensitive data transmission is significant, impacting the overall performance and security of the distributed simulation platform. Therefore, there is an urgent need for a flexible scheduling method and system that can solve the above problems and realize cross-domain dynamic dependency processing, extremely fast scheduling of real-time tasks and secure adaptation of hybrid cloud resources. Summary of the Invention

[0003] The purpose of the present invention is to solve at least one of the technical problems existing in the prior art and to provide a distributed simulation platform elastic scheduling method and system based on cloud computing, which can solve the above problem.

[0004] To achieve the above objectives, the present invention provides the following technical solutions: a distributed simulation platform elastic scheduling system based on cloud computing, comprising a cross-domain dynamic dependency processing module, a real-time task rapid scheduling module, a hybrid cloud resource adaptation module, and an elastic scheduling core engine; The cross-domain dynamic dependency processing module is used to realize dynamic processing of dependency relationships, cross-domain collaborative allocation of resources, and self-healing of task chain breakpoints in multi-domain collaborative simulation; The real-time task rapid scheduling module is used to achieve predictive resource preheating, microsecond-level scheduling response, and lossless preemption of high-priority tasks for real-time simulation tasks. A hybrid cloud resource adaptation module, which is used to implement hierarchical scheduling of private and public cloud resources based on the sensitivity of task data and hardware-accelerated encrypted transmission of sensitive data; The elastic scheduling core engine is connected to the cross-domain dynamic dependency processing module, the real-time task rapid scheduling module, and the hybrid cloud resource adaptation module respectively, and is used to coordinate the collaborative work of each module and dynamically adjust the scheduling strategy.

[0005] Preferably, the cross-domain dynamic dependency processing module includes: A semantic dependency parser that automatically analyzes the correlation between input and output parameters of multi-domain simulation tasks and generates a real-time updated dependency graph; A cross-domain resource pool collaborative scheduler is used to dynamically adjust resource quotas based on inter-domain dependencies, enabling cross-domain reuse of idle resources. The task chain self-healing controller is used to monitor the status of each link in the task chain formed by multi-domain collaboration, and trigger breakpoint rescheduling when a link is delayed.

[0006] Preferably, the semantic dependency parser includes a domain interface registry, a dependency graph generator, and a dynamic rule engine; The domain interface registry is used to store data interface information registered when each domain simulation task is started; The dependency graph generator is used to construct a dynamic dependency network based on the correlation of interface parameters, and the weight of the dynamic dependency network is adjusted in real time according to the task type; The dynamic rule engine is used to automatically update the data synchronization frequency when an emergency scenario triggers a change in data demand.

[0007] Preferably, the cross-domain resource pool collaborative scheduler includes a domain resource profile library, a dependency sensitivity calculator, and a resource elastic allocation algorithm; The domain resource portrait library is used to record resource load information of each domain in real time; The dependency sensitivity calculator is used to quantify the resource dependency intensity between domains; The resource elastic allocation algorithm is used to call computing power from idle resource pools in other areas when capacity expansion is required in a certain area, while retaining redundant resources required for core tasks in other areas.

[0008] Preferably, the task chain self-healing controller includes a link status monitoring probe, a delay conduction simulator and a breakpoint rescheduling engine; The link status monitoring probe is used to collect the completion progress of each link in the task chain; The delay conduction simulator is used to predict the impact of a certain link delay on subsequent links; The breakpoint rescheduling engine is used to start the backup resource pool to expand computing power when a certain link is delayed, and suspend subsequent links to wait until the breakpoint is restored and then advance synchronously.

[0009] Preferably, the real-time task rapid scheduling module includes: Predictive resource preheating pool, used to predict resource demand based on historical load data and prepare resources in advance; Microsecond scheduling engine for achieving sub-second response for real-time tasks; Priority preemption arbiter, used to achieve lossless preemption of low-priority tasks by high-priority tasks.

[0010] Preferably, the predictive resource preheating pool includes a historical load prediction model, a resource snapshot manager and a preheating node pool; The historical load prediction model is used to predict resource requirements in subsequent time periods based on historical load data; The resource snapshot manager is used to generate a common simulation environment image; The preheated node pool is used to start the standby nodes before the predicted peak and maintain the standby nodes in a sleep-wake state.

[0011] Preferably, the hybrid cloud resource adaptation module includes: A cloud resource hierarchical scheduler that dynamically selects private or public cloud resources based on the sensitivity of task data; Hardware-accelerated encryption gateway, used to implement hardware-level encryption of sensitive data during cross-domain transmission and calculation; The cloud resource hierarchical scheduler includes a data sensitivity label library, a cloud resource security level matrix, and a cross-cloud link optimizer. The data sensitivity label library labels task data according to its sensitivity, such as military deployment simulations as top secret and public crowd flow heat maps as public. The cloud resource security level matrix matches cloud resources based on data sensitivity. Top-secret tasks are permanently scheduled to nodes in the private cloud with hardware encryption cards, and cross-cloud transmission is prohibited. Public-level tasks are automatically assigned to elastic nodes in the public cloud, and can be temporarily expanded to edge nodes during peak periods. The cross-cloud link optimizer optimizes cross-cloud data transmission links to reduce transmission latency. The hardware-accelerated encryption gateway includes the SGX encryption chip, the National Encryption SM algorithm accelerator and the transport layer encryption tunnel manager. The SGX encryption chip provides a hardware-level security environment, the National Encryption SM algorithm accelerator improves encryption efficiency, and the transport layer encryption tunnel manager establishes an encrypted transmission tunnel, reducing the encryption time of sensitive data such as emergency plan parameters during cross-domain transmission and calculation to 1 / 5 of software encryption.

[0012] Preferably, the microsecond scheduling engine includes a load monitoring high-frequency collector, a resource status memory database, and an extremely fast decision maker; The load monitoring high-frequency collector is used to monitor the node load at high frequency and write the data into the resource status memory database in real time; The extreme speed decision maker is used to call resources directly from the preheating pool when a real-time task is detected to be triggered, and compress resource readiness time through kernel-level scheduling; The priority preemption arbiter includes a task priority register, a state saving buffer and a resource switching controller; The task priority registry is used to preset a task priority matrix; The state saving buffer is used to store the memory data of the low priority task in real time when the high priority task is triggered; The resource switching controller is used to release the occupied node resources and allocate them to the high-priority task, and after the high-priority task is completed, restore the low-priority task through the buffer data.

[0013] The distributed simulation platform elastic scheduling method based on cloud computing includes the following steps: S1. Cross-domain dynamic dependency processing step: Dynamically process the dependency relationships in multi-domain collaborative simulation to achieve cross-domain collaborative resource allocation and task chain breakpoint self-healing; The steps for handling cross-domain dynamic dependencies include: Semantic dependency parsing step: Automatically parse the input and output parameter correlations of multi-domain simulation tasks and generate a real-time updated dependency graph; Cross-domain resource pool collaborative scheduling steps: Dynamically adjust resource quotas based on inter-domain dependencies to achieve cross-domain reuse of idle resources; Task chain self-healing steps: monitor the status of each link in the task chain and trigger breakpoint rescheduling when a link is delayed; S2, real-time task rapid scheduling steps: predictive resource preheating, microsecond-level scheduling response, and lossless preemption of high-priority tasks for real-time simulation tasks; The real-time task rapid scheduling step includes: Predictive resource warm-up step: predict resource demand based on historical load data and prepare resources in advance; Microsecond scheduling step: achieve sub-second response for real-time tasks; S3, hybrid cloud resource adaptation step: hierarchical scheduling of private cloud and public cloud resources based on the sensitivity of task data, and hardware-accelerated encryption transmission of sensitive data; The hybrid cloud resource adaptation step includes: Cloud resource hierarchical scheduling steps: Dynamically select private cloud or public cloud resources based on the sensitivity of task data; Hardware-accelerated encryption transmission step: hardware-level encryption is performed on sensitive data during cross-domain transmission and calculation; S4, elastic scheduling core control step: coordinate the above steps to work together and dynamically adjust the scheduling strategy based on real-time monitoring data.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This cloud computing-based distributed simulation platform elastic scheduling method and system, through a cross-domain dynamic dependency processing module, realizes dynamic analysis of multi-domain dependency relationships, cross-domain collaborative allocation of resources and breakpoint self-healing of task chains, solves the problem of rigid cross-domain dependency processing mechanism in the existing technology, reduces cross-domain data synchronization delay, and improves resource utilization and task chain stability.

[0015] 2. The cloud computing-based distributed simulation platform elastic scheduling method and system, the real-time task rapid scheduling module significantly improves the scheduling response speed of real-time tasks through predictive resource preheating, microsecond-level scheduling response and lossless preemption of high-priority tasks, meets the timeliness requirements of real-time simulation tasks, and avoids the problems of delayed resource acquisition of high-priority tasks and loss of interrupted task status.

[0016] 3. The cloud computing-based distributed simulation platform elastic scheduling method and system, the hybrid cloud resource adaptation module realizes the hierarchical scheduling of private cloud and public cloud resources according to the sensitivity of task data, and encrypts the transmission of sensitive data through the hardware-accelerated encryption gateway. While ensuring data security, it reduces the encryption performance loss and solves the contradiction between security and performance in hybrid cloud resource adaptation.

[0017] 4. The elastic scheduling method and system of the distributed simulation platform based on cloud computing. The elastic scheduling core engine coordinates the collaborative work of various modules and dynamically adjusts the scheduling strategy, enabling the entire system to adaptively optimize according to real-time monitoring data, further enhancing the elastic scheduling capabilities of the distributed simulation platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a schematic diagram of the framework of the elastic scheduling system of the distributed simulation platform based on cloud computing of the present invention. DETAILED DESCRIPTION

[0019] This section will describe in detail the specific embodiments of the present invention. The preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the accompanying drawings is to supplement the description of the text part of the specification with graphics, so that people can intuitively and vividly understand each technical feature and the overall technical solution of the present invention, but they should not be understood as limiting the scope of protection of the present invention.

[0020] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, they cannot be understood as limitations on the present invention.

[0021] In the description of this invention, terms such as "greater than," "less than," and "exceed" are understood to exclude the number itself, while terms such as "above," "below," and "within" are understood to include the number itself. The use of terms such as "first" and "second" is solely for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly specifying the number of the indicated technical features, or implicitly specifying the order of the indicated technical features.

[0022] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, and connecting should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0023] See also Figure 1 The present invention provides a technical solution: a distributed simulation platform elastic scheduling system based on cloud computing, including a cross-domain dynamic dependency processing module, a real-time task rapid scheduling module, a hybrid cloud resource adaptation module and an elastic scheduling core engine. The elastic scheduling core engine is connected to the cross-domain dynamic dependency processing module, the real-time task rapid scheduling module and the hybrid cloud resource adaptation module respectively, and is used to coordinate the collaborative work of each module and dynamically adjust the scheduling strategy according to real-time monitoring data; The cross-domain dynamic dependency processing module includes a semantic dependency parser, a cross-domain resource pool collaborative scheduler, and a task chain self-healing controller; The semantic dependency parser includes a domain interface registry, a dependency graph generator, and a dynamic rule engine. In actual applications, when each domain simulation task is started, data interface information is registered with the domain interface registry, such as weather simulation → precipitation intensity → output frequency. The dependency graph generator builds a dynamic dependency network based on the correlation between these interface parameters, and the weight of the network is adjusted in real time according to the task type. When sudden scenarios such as traffic accidents trigger changes in data requirements, the dynamic rule engine automatically updates the synchronization frequency, for example, from 5 minutes / time to 100ms / time. The cross-domain resource pool collaborative scheduler includes a domain resource profile library, a dependency sensitivity calculator, and a resource elastic allocation algorithm. The domain resource profile library records resource load information for each domain in real time, such as GPU utilization in the energy domain and CPU occupancy in the transportation domain. The dependency sensitivity calculator quantifies the intensity of resource dependency between domains. For example, during heavy rainfall, the dependency of transportation on the weather domain increases to 0.8; When the transportation sector needs to expand CPU capacity by 30% due to increased precipitation intensity, the resource elasticity allocation algorithm automatically draws computing power from the idle resource pool in the energy sector, while retaining 20% ​​of redundant resources required for core tasks in the energy sector. The task chain self-healing controller includes a link status monitoring probe, a delay conduction simulator, and a breakpoint rescheduling engine. The link status monitoring probe collects the completion progress of each link in the task chain every second, such as whether the pollution diffusion simulation outputs the pollutant concentration distribution on time; The delay conduction simulator predicts the impact of a delay in a certain link on subsequent links. For example, a 200ms delay in pollution simulation will cause a 1.2-second delay in traffic control decisions. When a link is delayed, the breakpoint rescheduling engine immediately activates the backup resource pool, expands the computing power of the delayed link, and pauses subsequent links to wait until the breakpoint is recovered and the process is carried out synchronously. The real-time task rapid scheduling module includes a predictive resource preheating pool, a microsecond-level scheduling engine, and a priority preemptive arbiter; The predictive resource preheating pool includes a historical load prediction model, a resource snapshot manager, and a preheating node pool. The historical load prediction model uses the LSTM time series algorithm. Every morning, based on the previous day's load data, such as traffic simulation CPU utilization and GPU demand frequency, it predicts resource demand for each time period of the next day. The resource snapshot manager generates common simulation environment images, such as GPU node snapshots with smoke diffusion algorithms; The preheat node pool starts the standby node 30 minutes before the predicted peak, maintaining a sleep-wake state, reducing the startup time from 90 seconds to 500ms. The microsecond-level scheduling engine includes a high-frequency load monitoring collector, a resource status memory database, and an extremely fast decision maker. The high-frequency load monitoring collector monitors node loads such as CPU, memory, and network IO at a frequency of 10ms / time and writes the data to the resource status memory database in real time. When a real-time task such as traffic light timing simulation is detected, the rapid decision maker skips the traditional application-approval process and directly calls resources from the preheated pool. For containerized tasks, it uses kernel-level scheduling methods such as Kubernetes static Pod binding to reduce resource readiness time to less than 2 seconds. The priority preemption arbiter includes a task priority registry, a state saving buffer, and a resource switching controller. The task priority registry presets a task priority matrix, such as P0 for earthquake emergency and P3 for daily simulation. When a P0-level task is triggered, the state save buffer stores the memory data of the P3-level task in real time within ≤100ms; The resource switching controller releases the preempted node resources and allocates them to the P0-level task. After the high-priority task is completed, the P3-level task is restored through the buffer data without recalculation.

[0024] The hybrid cloud resource adaptation module includes a cloud resource hierarchical scheduler and a hardware-accelerated encryption gateway; The cloud resource hierarchical scheduler includes a data sensitivity label library, a cloud resource security level matrix, and a cross-cloud link optimizer. The data sensitivity label library labels task data according to its sensitivity, such as military deployment simulations as top secret and public crowd flow heat maps as public. The cloud resource security level matrix matches cloud resources based on data sensitivity. Top-secret tasks are permanently scheduled to nodes in the private cloud with hardware encryption cards, and cross-cloud transmission is prohibited. Public-level tasks are automatically assigned to elastic nodes in the public cloud, and can be temporarily expanded to edge nodes during peak periods. The cross-cloud link optimizer optimizes cross-cloud data transmission links to reduce transmission latency. The hardware-accelerated encryption gateway includes the SGX encryption chip, the National Encryption SM algorithm accelerator and the transport layer encryption tunnel manager. The SGX encryption chip provides a hardware-level security environment, the National Encryption SM algorithm accelerator improves encryption efficiency, and the transport layer encryption tunnel manager establishes an encrypted transmission tunnel, reducing the encryption time of sensitive data such as emergency plan parameters during cross-domain transmission and calculation to 1 / 5 of software encryption.

[0025] An embodiment of the present invention further provides a distributed simulation platform elastic scheduling method based on cloud computing, comprising the following steps: S1. Cross-domain dynamic dependency processing steps: S11, semantic dependency parsing step: register the data interface when each domain task starts, build a dynamic dependency network based on the correlation of interface parameters, and automatically update the data synchronization frequency in emergency scenarios; S12, cross-domain resource pool collaborative scheduling step: Real-time recording of resource loads in each domain, quantification of resource dependency between domains, and when capacity expansion is needed in a certain domain, computational power can be transferred from idle resource pools in other domains; S13, Task chain self-healing step: Collect the completion progress of each link in the task chain, predict the impact of delays, and start the backup resource pool expansion when a link is delayed and suspend subsequent links to wait; S2. Real-time task rapid scheduling steps: S21, predictive resource warm-up step: predict resource demand based on historical load data, generate a simulation environment image, start the standby node before the predicted peak and keep it in a "sleep-wake" state; S22, microsecond-level scheduling step: Monitor node loads at high frequency and store data in real time. When real-time tasks are triggered, resources are directly called from the preheating pool, shortening resource readiness time through kernel-level scheduling. S23, priority preemption step: preset task priority matrix, store low-priority task memory data when high-priority task is triggered, release resources to allocate to high-priority task, and resume low-priority task after completion; S3, hybrid cloud resource adaptation steps: S31, cloud resource hierarchical scheduling step: sensitivity tagging of task data, and matching of private cloud or public cloud resources based on sensitivity; S32, hardware accelerated encryption transmission step: performing hardware-level encryption transmission on sensitive data through a hardware encryption chip and algorithm accelerator; S4. Elastic scheduling core control steps: Coordinate the above steps to work together. When the cross-domain synchronization delay is detected to be greater than 500ms, the graph update frequency of the semantic parser is automatically increased. When the high-priority task preemption success rate is less than 100%, the priority registry rule recalibration is triggered.

[0026] Taking the rainstorm weather → traffic-emergency collaborative simulation scenario as an example, the working process of the present invention is explained: Dynamic dependency activation: When a "heavy rainfall" warning is triggered in the weather domain, the semantic dependency parser adjusts the synchronization frequency of weather data in the transportation domain from 5 minutes to 100ms, and the dependency graph weights are updated in real time. Resource elastic adaptation: The cross-domain resource pool collaborative scheduler detected a 30% surge in CPU demand in the transportation sector and allocated 20% of computing power from the idle resource pool in the energy sector to preserve redundancy for core energy tasks while simultaneously starting up GPU nodes in the preheating pool as backup. Real-time task response: A sudden traffic accident triggered a P0-level emergency simulation. The ultra-fast scheduling engine woke up two GPU nodes from the preheating pool within 1.2 seconds. The priority arbiter saved the low-priority energy-consuming simulation state in 80ms. Resources were switched to the emergency simulation task, and the hardware encryption gateway was simultaneously activated to protect the accident scene data. Task chain self-healing: If the pollution diffusion simulation is delayed by 150ms due to resource fluctuations, the task chain self-healing controller will immediately suspend subsequent traffic control simulations, start backup node expansion, and restore the pollution simulation progress within 200ms to ensure that public warnings are issued without delay.

[0027] The embodiments of the present invention are described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Various changes can be made within the scope of knowledge possessed by ordinary technicians in the technical field without departing from the scope of the present invention.

Claims

1. A cloud computing-based distributed simulation platform elastic scheduling system, including a cross-domain dynamic dependency processing module, a real-time task rapid scheduling module, a hybrid cloud resource adaptation module, and an elastic scheduling core engine; Its characteristics are: The cross-domain dynamic dependency processing module is used to realize dynamic processing of dependency relationships, cross-domain collaborative allocation of resources, and self-healing of task chain breakpoints in multi-domain collaborative simulation; The real-time task rapid scheduling module is used to achieve predictive resource preheating, microsecond-level scheduling response, and lossless preemption of high-priority tasks for real-time simulation tasks. A hybrid cloud resource adaptation module, which is used to implement hierarchical scheduling of private and public cloud resources based on the sensitivity of task data and hardware-accelerated encrypted transmission of sensitive data; The elastic scheduling core engine is connected to the cross-domain dynamic dependency processing module, the real-time task rapid scheduling module, and the hybrid cloud resource adaptation module respectively, and is used to coordinate the collaborative work of each module and dynamically adjust the scheduling strategy.

2. The distributed simulation platform elastic scheduling system based on cloud computing according to claim 1, characterized in that: The cross-domain dynamic dependency processing module includes: A semantic dependency parser that automatically analyzes the correlation between input and output parameters of multi-domain simulation tasks and generates a real-time updated dependency graph; A cross-domain resource pool collaborative scheduler is used to dynamically adjust resource quotas based on inter-domain dependencies, enabling cross-domain reuse of idle resources. The task chain self-healing controller is used to monitor the status of each link in the task chain formed by multi-domain collaboration, and trigger breakpoint rescheduling when a link is delayed.

3. The distributed simulation platform elastic scheduling system based on cloud computing according to claim 2, characterized in that: The semantic dependency parser includes a domain interface registry, a dependency graph generator, and a dynamic rule engine; The domain interface registry is used to store data interface information registered when each domain simulation task is started; The dependency graph generator is used to construct a dynamic dependency network based on the correlation of interface parameters, and the weight of the dynamic dependency network is adjusted in real time according to the task type; The dynamic rule engine is used to automatically update the data synchronization frequency when an emergency scenario triggers a change in data demand.

4. The distributed simulation platform elastic scheduling system based on cloud computing according to claim 3, characterized in that: The cross-domain resource pool collaborative scheduler includes a domain resource profile library, a dependency sensitivity calculator, and a resource elastic allocation algorithm; The domain resource portrait library is used to record resource load information of each domain in real time; The dependency sensitivity calculator is used to quantify the resource dependency intensity between domains; The resource elastic allocation algorithm is used to call computing power from idle resource pools in other areas when capacity expansion is required in a certain area, while retaining redundant resources required for core tasks in other areas.

5. The distributed simulation platform elastic scheduling system based on cloud computing according to claim 4 is characterized in that: The task chain self-healing controller includes a link status monitoring probe, a delay conduction simulator and a breakpoint rescheduling engine; The link status monitoring probe is used to collect the completion progress of each link in the task chain; The delay conduction simulator is used to predict the impact of a certain link delay on subsequent links; The breakpoint rescheduling engine is used to start the backup resource pool to expand computing power when a certain link is delayed, and suspend subsequent links to wait until the breakpoint is restored and then advance synchronously.

6. The distributed simulation platform elastic scheduling system based on cloud computing according to claim 5, characterized in that: The real-time task rapid scheduling module includes: Predictive resource preheating pool, used to predict resource demand based on historical load data and prepare resources in advance; Microsecond scheduling engine for achieving sub-second response for real-time tasks; Priority preemption arbiter, used to achieve lossless preemption of low-priority tasks by high-priority tasks.

7. The distributed simulation platform elastic scheduling system based on cloud computing according to claim 6, characterized in that: The predictive resource preheating pool includes a historical load prediction model, a resource snapshot manager and a preheating node pool; The historical load prediction model is used to predict resource requirements in subsequent time periods based on historical load data; The resource snapshot manager is used to generate a common simulation environment image; The preheated node pool is used to start the standby nodes before the predicted peak and maintain the standby nodes in a sleep-wake state.

8. The distributed simulation platform elastic scheduling system based on cloud computing according to claim 7, characterized in that: The hybrid cloud resource adaptation module includes: A cloud resource hierarchical scheduler that dynamically selects private or public cloud resources based on the sensitivity of task data; Hardware-accelerated encryption gateway, used to implement hardware-level encryption of sensitive data during cross-domain transmission and calculation; The cloud resource hierarchical scheduler includes a data sensitivity label library, a cloud resource security level matrix, and a cross-cloud link optimizer. The data sensitivity label library labels task data according to its sensitivity, such as military deployment simulations as top secret and public crowd flow heat maps as public. The cloud resource security level matrix matches cloud resources based on data sensitivity. Top-secret tasks are permanently scheduled to nodes in the private cloud with hardware encryption cards, and cross-cloud transmission is prohibited. Public-level tasks are automatically assigned to elastic nodes in the public cloud, and can be temporarily expanded to edge nodes during peak periods. The cross-cloud link optimizer optimizes cross-cloud data transmission links to reduce transmission latency. The hardware-accelerated encryption gateway includes the SGX encryption chip, the National Encryption SM algorithm accelerator and the transport layer encryption tunnel manager. The SGX encryption chip provides a hardware-level security environment, the National Encryption SM algorithm accelerator improves encryption efficiency, and the transport layer encryption tunnel manager establishes an encrypted transmission tunnel, reducing the encryption time of sensitive data such as emergency plan parameters during cross-domain transmission and calculation to 1 / 5 of software encryption.

9. The distributed simulation platform elastic scheduling system based on cloud computing according to claim 8, characterized in that: The microsecond scheduling engine includes a load monitoring high-frequency collector, a resource status memory database, and an extremely fast decision maker; The load monitoring high-frequency collector is used to monitor the node load at high frequency and write the data into the resource status memory database in real time; The extreme speed decision maker is used to call resources directly from the preheating pool when a real-time task is detected to be triggered, and compress resource readiness time through kernel-level scheduling; The priority preemption arbiter includes a task priority register, a state saving buffer and a resource switching controller; The task priority registry is used to preset a task priority matrix; The state saving buffer is used to store the memory data of the low priority task in real time when the high priority task is triggered; The resource switching controller is used to release the occupied node resources and allocate them to the high-priority task, and after the high-priority task is completed, restore the low-priority task through the buffer data.

10. A distributed simulation platform elastic scheduling method based on cloud computing, characterized by: The following steps are involved: S1. Cross-domain dynamic dependency processing step: Dynamically process the dependency relationships in multi-domain collaborative simulation to achieve cross-domain collaborative resource allocation and task chain breakpoint self-healing; The steps for handling cross-domain dynamic dependencies include: Semantic dependency parsing step: Automatically parse the input and output parameter correlations of multi-domain simulation tasks and generate a real-time updated dependency graph; Cross-domain resource pool collaborative scheduling steps: Dynamically adjust resource quotas based on inter-domain dependencies to achieve cross-domain reuse of idle resources; Task chain self-healing steps: monitor the status of each link in the task chain and trigger breakpoint rescheduling when a link is delayed; S2, real-time task rapid scheduling steps: predictive resource preheating, microsecond-level scheduling response, and lossless preemption of high-priority tasks for real-time simulation tasks; The real-time task rapid scheduling step includes: Predictive resource warm-up step: predict resource demand based on historical load data and prepare resources in advance; Microsecond scheduling step: achieve sub-second response for real-time tasks; S3, hybrid cloud resource adaptation step: hierarchical scheduling of private cloud and public cloud resources based on the sensitivity of task data, and hardware-accelerated encryption transmission of sensitive data; The hybrid cloud resource adaptation step includes: Cloud resource hierarchical scheduling steps: Dynamically select private cloud or public cloud resources based on the sensitivity of task data; Hardware-accelerated encryption transmission step: hardware-level encryption is performed on sensitive data during cross-domain transmission and calculation; S4, elastic scheduling core control step: coordinate the above steps to work together and dynamically adjust the scheduling strategy based on real-time monitoring data.

Images