Cross-regional data collaborative computing method and system based on multi-level cache mechanism

Abstract

The present application discloses a cross-regional data collaborative computing method and system based on a multi-level cache mechanism, which relates to the field of cloud computing technology. The method includes: obtaining the historical operating status sequence of each cloud cache platform and the task description data of the target computing task to be scheduled; inputting the task description data and the operating status data of each cloud cache platform into the platform scheduling model to output a matching cross-regional data collaborative path; packaging the target computing task and its dependent environment to generate a corresponding task container image, and sending the task container image to each potential cloud cache platform; controlling the first potential cloud cache platform to run the corresponding task container image, monitoring the real-time operating status of the first potential cloud cache platform, and switching to control the second potential cloud cache platform to run the corresponding task container image when the monitored real-time operating status meets the task migration condition. In this way, task migration scheduling and resource management based on the multi-level cache mechanism are realized.

Description

Technical Field

[0001] The present application relates to the field of cloud computing technology, and in particular to a cross-regional data collaborative computing method and system based on a multi-level cache mechanism. Background Art

[0002] The "Eastern Data, Western Computing" initiative is a national strategy aimed at optimizing and synergizing the computing capabilities of data centers in eastern China with those in western China. With the development of cloud computing and big data, collaborative computing between data centers can effectively improve resource utilization and reduce latency.

[0003] Currently, data center resource scheduling and data storage management play a crucial role in cross-regional data collaborative computing. In large-scale distributed computing environments, data storage and access efficiency are crucial to computing performance. While existing distributed computing and storage solutions have alleviated the pressure of cross-regional data processing to some extent, multi-level cache systems often fail to effectively coordinate with task scheduling mechanisms in actual deployments.

[0004] Furthermore, existing data center resource scheduling strategies are primarily based on static or rule-based scheduling algorithms, which are unable to adapt to the dynamic changes in cloud platform resource status in real time. This is especially true in cross-region collaborative computing scenarios, where both computing and cache resource utilization fluctuate significantly. Traditional scheduling strategies struggle to optimize resource allocation in response to changes in resource load and sudden task demands, leading to significant task delays and resource waste. Summary of the Invention

[0005] The present application provides a cross-regional data collaborative computing method, system, storage medium, computer program product and electronic device based on a multi-level caching mechanism, which is used to at least solve the problems of network transmission bottlenecks, insufficient caching strategies, static scheduling decisions and limited overall collaborative capabilities of traditional multiple cloud cache platform resources in cross-regional data collaboration in current related technologies.

[0006] In the first aspect, the embodiment of the present application provides a cross-regional data collaborative computing method based on a multi-level cache mechanism, including: obtaining the historical operating status sequence of each cloud cache platform and the task description data of the target computing task to be collaboratively calculated; the historical operating status sequence includes multiple historical time windows and corresponding historical operating status data, and the parameter types of the historical operating status data include cache hit rate, cache delay, cache capacity utilization, computing resource utilization and network transmission delay; the task description data includes task data size, task data cache requirement, task computing resource requirement, task type and task processing time limit; the task description data and the historical operating status sequence of each cloud cache platform are input into the platform scheduling model to output A matching cross-regional data collaboration path; the cross-regional data collaboration path defines multiple potential cloud cache platforms arranged in descending order of priority; the platform scheduling model is a deep learning model; the target computing task and its dependent environment are packaged to generate a corresponding task container image, and the task container image is sent to each of the potential cloud cache platforms; the first potential cache cloud cache platform is controlled to run the corresponding task container image, the real-time operating status of the first potential cloud cache platform is monitored, and when the monitored real-time operating status meets the task migration conditions, the second potential cloud cache platform is switched to control the operation of the corresponding task container image; the priority of the first potential cloud cache platform is higher than that of the second potential cache cloud cache platform.

[0007] In the second aspect, the embodiment of the present application provides a cross-regional data collaborative computing system based on a multi-level cache mechanism, including: a data acquisition unit, used to obtain the historical operating status sequence of each cloud cache platform and the task description data of the target computing task to be collaboratively calculated; the historical operating status sequence includes multiple historical time windows and corresponding historical operating status data, and the parameter types of the historical operating status data include cache hit rate, cache delay, cache capacity utilization, computing resource utilization and network transmission delay; the task description data includes task data size, task data cache requirement, task computing resource requirement, task type and task processing time limit; a collaborative path matching unit, used to input the task description data and the historical operating status sequence of each of the cloud cache platforms into the platform scheduling model to output the corresponding A matching cross-regional data collaboration path; the cross-regional data collaboration path defines multiple potential cloud cache platforms arranged in descending order of priority; the platform scheduling model is a deep learning model; a task image synchronization unit is used to package the target computing task and its dependent environment to generate a corresponding task container image, and send the task container image to each of the potential cloud cache platforms; a platform monitoring and switching unit is used to control the first potential cache cloud cache platform to run the corresponding task container image, monitor the real-time operating status of the first potential cloud cache platform, and switch to control the second potential cloud cache platform to run the corresponding task container image when the monitored real-time operating status meets the task migration conditions; the priority of the first potential cloud cache platform is higher than that of the second potential cache cloud cache platform.

[0008] According to a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the steps of the cross-regional data collaborative computing method based on a multi-level cache mechanism of any embodiment of the present application.

[0009] In a fourth aspect, an embodiment of the present application provides a storage medium on which a computer program is stored, characterized in that when the program is executed by a processor, the steps of the cross-regional data collaborative computing method based on a multi-level cache mechanism of any embodiment of the present application are implemented.

[0010] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the cross-regional data collaborative computing method based on a multi-level cache mechanism of any embodiment of the present application.

[0011] The cross-regional data collaborative computing method and system based on a multi-level cache mechanism provided by this application can produce at least the following technical effects:

[0012] (1) Through the deep learning model, the historical operating status and target task data of multiple cloud cache platforms are analyzed. With dynamic learning and adaptive capabilities, it can output the optimal cross-regional data collaboration path based on the historical and real-time data of task type, computing resource requirements and platform bandwidth. Based on the intelligent matching of cloud cache platform resources and data computing tasks, it realizes the automatic coordination of resources and dynamic migration of tasks. By monitoring the operating status of the cloud cache platform in real time, during the task execution process, once it is found that the resource usage of the current cloud cache platform exceeds the threshold or no longer meets the task requirements, it can trigger the call of other backup cloud cache platforms with lower priority according to the cross-regional data collaboration path to continue the task. Therefore, the migration mechanism based on real-time status monitoring ensures the successful response rate and reliability of the task.

[0013] (2) Encapsulating the target task and its dependent environment as a container image makes the migration of tasks between different cloud cache platforms fast and seamless. Containerization technology not only ensures the portability of data computing tasks, but also makes switching between platforms almost ineffective on task execution. With the support of standardized container image packaging, both edge computing platforms and public cloud cache platforms can execute tasks in a standardized manner, ensuring the continuity of cross-platform collaboration.

[0014] Through this technical solution, by integrating deep learning models, containerization technology and cross-platform resource dynamic migration mechanism, task migration scheduling and resource management based on multi-level caching mechanism are realized, which realizes intelligent scheduling and resource management in multi-cloud and hybrid cloud environments, and can adaptively respond to changes in complex dynamic environments, thereby improving resource utilization and the continuity of task execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0016] Figure 1 A flowchart illustrating an example of a cross-region data collaborative computing method based on a multi-level cache mechanism according to an embodiment of the present application is shown;

[0017] Figure 2 Shown according to Figure 1 An exemplary operational flow chart of step S130 in FIG.

[0018] Figure 3 Shown according to Figure 2An exemplary operational flow chart of step S230 in FIG.

[0019] Figure 4 A schematic diagram showing a structural connection of an example of a platform scheduling model according to an embodiment of the present application is shown;

[0020] Figure 5 A structural block diagram of an example of a cross-regional data collaborative computing system based on a multi-level cache mechanism according to an embodiment of the present application is shown;

[0021] Figure 6 This is a schematic structural diagram of an embodiment of an electronic device of the present application. DETAILED DESCRIPTION

[0022] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0023] In the technical solutions of this application, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved shall comply with the provisions of relevant laws and regulations and shall not violate public order and good morals.

[0024] Figure 1 A flowchart of an example of a cross-region data collaborative computing method based on a multi-level cache mechanism according to an embodiment of the present application is shown.

[0025] The execution entity of the method of the embodiment of the present application can be any controller or processor with computing or processing capabilities. Specifically, it can be implemented by the hybrid cloud cache platform scheduling platform or scheduling server. The automated task allocation and migration mechanism driven by deep learning models reduces the complexity of manual management, lowers system operating costs, and meets the needs of enterprises and scientific research projects for efficient cloud computing platforms.

[0026] In some examples, it can be integrated into an electronic device or terminal through software, hardware, or a combination of software and hardware, and the type of terminal or electronic device can be diverse, such as a mobile phone, tablet computer, or desktop computer, etc.

[0027] like Figure 1 As shown, in step S110, the historical operation status sequence of each cloud cache platform and the task description data of the target computing task to be collaboratively calculated are obtained.

[0028] Here, the historical operation status sequence includes multiple historical time windows and corresponding historical operation status data, and the parameter types of the historical operation status data include cache hit rate, cache delay, cache capacity utilization, computing resource utilization and network transmission delay. In some embodiments, an API or log analysis tool (such as Prometheus or CloudWatch, etc.) is used to obtain historical monitoring data of each cloud cache platform, and the data is divided according to different time windows (such as hourly, daily), and archived as parameters such as computing resource utilization (such as CPU, GPU, memory occupancy), network delay and bandwidth utilization (upload / download speed and bandwidth usage). The cache hit rate refers to the ratio of the number of requests to obtain data directly from the cache to the total number of requests in the cache request, reflecting the effectiveness of the cache. It can be automatically recorded and updated by the cache management agent deployed by the cloud cache platform. The cache delay refers to the time required from initiating a request to successfully obtaining data from the cache. The lower the delay, the higher the data access efficiency. For example, the response time of each request is recorded by the $request_time variable in Nginx. Cache capacity utilization indicates the ratio of the used storage space of the cache platform to the total cache capacity, reflecting the usage of cache resources. For example, you can view it through Redis commands.

[0029] The task description data includes the task data size, task computing resource requirements, task type, and task processing time limit. In some embodiments, after the access request reaches the hybrid cloud cache platform scheduling platform, the task data is extracted from the access request or scheduling task queue, specifically including the task data size (MB / GB) used to describe the volume of transmitted data, the task computing resource requirements (CPU / GPU number, memory requirements, etc.), the task data cache requirements, the task type (real-time task, batch task, etc.) and the time constraint for the task to be completed. Here, the task data cache requirement represents the target computing task's demand for cache space, which depends on the task's data scale, data access frequency, and data sharing degree, and can be distinguished by the task description analysis module, for example, divided into cache requirement levels of "low demand", "medium demand", and "high demand".

[0030] In step S120, the task description data and the historical operating status sequence of each cloud cache platform are input into the platform scheduling model to output a matching cross-regional data collaboration path.

[0031] Here, the platform scheduling model is a deep learning model. Specifically, the platform scheduling model can be selected from various non-restrictive time series models, such as LSTM (Long Short-Term Memory) and Transformer, to predict the operating status of each platform in the future and then output cross-regional data collaboration paths.

[0032] The cross-regional data collaboration path defines multiple potential cloud cache platforms ranked by priority from high to low. In some embodiments, the path is a list of cloud cache platforms ranked by priority, with the priority of each cloud cache platform calculated based on comprehensive factors such as its resource utilization, latency, and bandwidth. For example, the first potential cloud cache platform with the highest priority is the one whose expected operating status best meets the task requirements, while the other second potential cloud cache platforms serve as backup nodes and switch to operation when the first potential cloud cache platform has insufficient resources or anomalies.

[0033] In step S130 , the target computing task and its dependent environment are packaged to generate a corresponding task container image, and the task container image is sent to each potential cloud cache platform.

[0034] It should be noted that in traditional container management technology, application services are generally imaged as containers. For example, a long-running application service (such as a web server, database, API service, etc.) is packaged into a container image for cross-platform deployment and management. In contrast, in the embodiments of the present application, specific computing tasks (such as data analysis, batch processing tasks, etc.) are packaged into container images for execution across cloud cache platforms, and resources are released after the task is completed. Unlike the continuous operation of application services, tasks are usually short-term calculations that stop upon completion (such as a period of data analysis or a round of model training).

[0035] Table 1 Comparison between traditional application service container imaging and task container imaging in this embodiment

[0036]

[0037]

[0038] As shown in Table 1, application service mirroring favors stable operation and continuous resource utilization, while task mirroring focuses more on flexible resource usage and on-demand scheduling. Specifically, task container mirroring supports dynamic task migration, selecting the optimal platform based on real-time operating status, improving system elasticity and efficiency. Furthermore, task containers are destroyed upon completion, freeing up resources and avoiding prolonged cloud resource usage.

[0039] In particular, in business application scenarios, traditional application service container mirroring is suitable for running long-term and highly concurrent services, such as API gateways or databases. Regarding the task container mirroring in the embodiments of this application, it is more suitable for temporary tasks that require rapid completion of calculations, such as machine learning model training or batch data analysis. Therefore, through dynamic mirroring and cross-platform scheduling of tasks, resource utilization efficiency and system flexibility are improved.

[0040] In some embodiments, the task code, configuration files, and runtime environment (such as Python environment, TensorFlow and other dependent libraries) are encapsulated into a Docker image or other standardized container image. Then, the generated task container image is pushed to each potential cloud cache platform using the image transfer protocol of the distributed image repository or cloud cache platform. Preferably, the image size and transmission time can be optimized according to the data scale of the task or the cloud cache platform. For example, for a cloud cache platform with a large regional span, data compression or edge caching technology can be used to reduce transmission time.

[0041] Through the embodiments of this application, containerization technology is used to encapsulate the task's operating environment and dependencies, ensuring that tasks can be quickly deployed and run across different cloud cache platforms, reducing startup time. Furthermore, standardized container technology ensures high task portability and cross-platform compatibility, avoiding execution errors caused by environmental differences between different platforms and improving the robustness of the task scheduling system.

[0042] In step S140, the first potential cloud cache platform is controlled to run the corresponding task container image, the real-time running status of the first potential cloud cache platform is monitored, and when the monitored real-time running status meets the task migration condition, the second potential cloud cache platform is switched to control the running of the corresponding task container image. The priority of the first potential cloud cache platform is higher than that of the second potential cloud cache platform.

[0043] In some embodiments, tools such as Kubernetes or Prometheus are used to monitor resource usage during task execution, including CPU, memory, network latency, etc., to determine whether task migration conditions are triggered. For the description of task migration conditions, for example, they include resource overload, increased network latency, cost optimization requirements, etc. For example, when the computing resource utilization rate continues to be higher than the set threshold (such as 90%) for more than a period of time, or the platform's network transmission delay exceeds an acceptable range, or a cloud cache platform with lower costs is selected to run the task according to a preset strategy, etc. Furthermore, when the triggering condition for task migration is detected, the execution status of the task is directly restored on the new platform through the task mirror, avoiding task restart and improving migration efficiency.

[0044] Through the embodiments of the present application, during the task execution process, the optimal cloud cache platform is dynamically selected according to the changes in the real-time operating status. Based on the real-time monitoring and automated migration mechanism, the scheduling platform can respond to emergencies, ensuring seamless migration of tasks between different platforms, thereby improving the continuity of tasks and the fault tolerance of the system.

[0045] In some examples of the embodiments of the present application, when a task is migrated, the scheduling platform will suspend the execution of the current task, and transfer the task image and intermediate state to a second potential cloud cache platform that relays the task execution to ensure the continuity and performance stability of the task.

[0046] Specifically, while the first potential cloud cache platform is running the task container image, it generates a task checkpoint and synchronizes the task checkpoints between the various potential cloud cache platforms. The task checkpoints are used to store the intermediate execution status data of the target computing task. When the monitored first running status data meets the task migration conditions, the second potential cloud cache platform is switched to run the corresponding task container image according to the task checkpoint.

[0047] In some embodiments, a checkpoint file is used to save the intermediate execution status data of the task, which can be generated periodically, for example, a checkpoint is generated regularly every 10 minutes or every 1,000 data processing batches. Furthermore, the checkpoint file is stored in a distributed storage system (such as Ceph, HDFS) or cloud storage (such as AWS S3, Alibaba Cloud OSS, etc.) to ensure that all potential cloud cache platforms can access the latest task status. Incremental synchronization is performed between multiple potential cloud cache platforms, and only the data that has changed since the last checkpoint is transmitted to reduce the network overhead caused by synchronization. After the migration condition is triggered, the task on the first potential cloud cache platform is suspended to ensure that its execution status is consistent with the latest checkpoint. Furthermore, on the second potential cloud cache platform, the checkpoint file is loaded from the distributed storage, and the execution status of the task is restored to ensure that the task can continue to run seamlessly from the state before migration.

[0048] Through the embodiments of the present application, the complete intermediate execution state data is saved based on the task checkpoint, so that the task can continue to execute from the last saved state after migration, avoiding task restart or repeated calculation. The time the task is suspended during the migration process is greatly shortened, and the task continuity of the system is improved. Therefore, when the operating state changes, the system can quickly switch to other cloud cache platforms to continue executing tasks, improving the elasticity and robustness of the system. In addition, through the optimization of incremental synchronization and distributed storage, the checkpoint mechanism reduces the network transmission overhead during the migration process and ensures the efficient execution of tasks.

[0049] Regarding the implementation details of the above-mentioned step S130, in some embodiments, before generating the container image, the system will automatically analyze the dependency environment of the target task, which includes: the code and execution script of the computing task, the required system libraries (such as specific versions of Python, Java, etc.), the data dependencies of the initial data set required for task execution (such as configuration files or model parameters), and hardware requirements (such as GPU, number of CPU cores and memory size).

[0050] Then, use Docker or an OCI (Open Container Initiative)-compatible container imaging tool to package the code, dependency packages, and environment configuration into a standardized image file. Specifically, in the image build file (such as Dockerfile), package the task environment in stages to minimize the image size. In the first stage, select a base image (such as ubuntu:20.04, python:3.9-slim, etc.). In the second stage, install the task dependency packages (such as numpy, tensorflow, etc.). In the third stage, add the task code and required data.

[0051] For frequently used libraries and dependencies, a common base image layer (Base Image) is built to avoid rebuilding from scratch for each task, reducing build time and enabling incremental image updates. This way, when the task dependency environment changes, only the newly added layer of the image is updated instead of the entire image being rebuilt.

[0052] In some preferred implementations, image signing and encryption are used to ensure the security of images when transferred between different cloud caching platforms. Specifically, images are signed using tools such as Docker Content Trust to ensure their integrity. Furthermore, sensitive data layers are encrypted before image transfer to prevent data leakage.

[0053] Figure 2 Shown according to Figure 1 An exemplary operational flow chart of step S130 in FIG.

[0054] like Figure 2 As shown, in step S210, the task code, dependency package and environment configuration corresponding to the target computing task are parsed.

[0055] In some embodiments, a static code analysis tool (such as PyLint) is used to automatically parse the task code and identify dependencies. Library and module references in the task code are extracted to generate a dependency list. Exemplarily, the dependency management tool can use "npm" or "Maven" to capture libraries and versions.

[0056] Additionally, record system environment information, such as the operating system version, required runtime libraries, and Python version. Then, generate a configuration file (such as "requirements.txt" or "environment.yml") to describe the runtime environment required for the task. Use tools such as "conda" or "pip" to automatically generate dependency files.

[0057] In step S220 , a container image tool is called to package the parsed task code, dependency packages, and environment configuration to generate a corresponding initial image file.

[0058] For example, write a "Dockerfile" to specify the base image and build steps. Specifically, select a suitable base image and add instructions in the Dockerfile to install the parsed dependency packages. The relevant code snippet is:

[0059] ```Dockerfile

[0060] FROM python:3.9-slim

[0061] COPY requirements.txt.

[0062] RUN pip install-r requirements.txt

[0063] COPY . / app

[0064] WORKDIR / app

[0065] CMD["python","task.py"]

[0066] ```

[0067] Then, use staged build techniques to reduce the size of the final image. Specifically, use tools such as docker-slim to analyze and remove unnecessary files and layers to optimize the image. This reduces the image size, shortens pull times, and lowers storage costs.

[0068] In step S230 , the initial image file is encrypted to obtain a corresponding task container image, and the task container image is sent to each potential cloud cache platform.

[0069] It should be noted that encryption methods can be diverse, and encryption tools (such as "OpenSSL") can be used to encrypt the initial image file to ensure the security of data transmission. By adopting a container image signing mechanism, the integrity of the image during transmission is ensured. In some embodiments, the encrypted image file is pushed to the image repository of the cloud cache platform, and encryption and signing mechanisms can be used to ensure data security, enabling rapid transmission and storage of images between different cloud cache platforms.

[0070] Through the embodiments of the present application, task dependencies are automatically parsed and standardized image processing is performed to ensure seamless task execution across different cloud cache platforms. The security of the image during transmission is guaranteed through encryption and signing.

[0071] Figure 3 Shown according to Figure 2An example operation flow chart of step S230 in FIG.

[0072] like Figure 3 As shown, in step S310, the base layer data and application layer data in the initial image file are parsed, and the random session key and cloud cache platform public key for each potential cloud cache platform are obtained.

[0073] Specifically, the initial image file includes the task's dependency environment and execution logic, organized in a multi-layer structure, containing base layer data and application layer data. For example, base layer data may include the operating system, basic dependency packages (such as the Alpine Linux base image for containers), and common library files. Application layer data may include the task's business logic, specialized dependencies (such as deep learning models and database connection configurations), task code, and runtime environment.

[0074] In addition, the scheduling platform can generate a unique random session key for each potential cloud cache platform to ensure the independence of each platform's data transmission and access process. The corresponding cloud cache platform public key (the public key part of the asymmetric key) is obtained from the key management system of each potential cloud cache platform (such as AWS KMS, Alibaba Cloud KMS).

[0075] This separation of base-layer data from application-layer data allows image files to be loaded on demand across different cloud cache platforms, reducing redundant data transmission and improving transmission efficiency. Furthermore, a separate random session key is generated for each cloud cache platform and used with the key provided by its public key system, ensuring data security and preventing the risk of key leakage or reuse.

[0076] In step S320, the application layer data is symmetrically encrypted according to each random session key to generate corresponding encrypted application layer data.

[0077] In some embodiments, application-layer data is symmetrically encrypted (e.g., AES-256) using a random session key. The AES algorithm provides efficient and powerful encryption, ensuring that application-layer data is protected from third-party theft during transmission. Preferably, large-scale application-layer data (e.g., large model files) is encrypted in blocks, and each data block is encrypted in parallel using multithreading or GPU acceleration to increase encryption speed.

[0078] In step S330, for each potential cloud cache platform, the corresponding random session key is asymmetrically encrypted using the cloud cache platform public key of the potential cloud cache platform to obtain the corresponding encrypted session key, and the basic layer data, encrypted application layer data and encrypted session key are packaged into a corresponding encrypted transmission package.

[0079] Asymmetric encryption algorithms, such as RSA, can be used to asymmetrically encrypt random session keys, ensuring they are not leaked during transmission. Each encrypted session key is unique to the cloud cache platform, and other platforms cannot decrypt it using their own private keys, thus ensuring data access control.

[0080] Furthermore, the base layer data (unencrypted), the encrypted application layer data, and the encrypted session key are integrated into an encrypted transmission package. By adopting a standardized data format (such as JSON or Protobuf), the transmission package can be smoothly parsed and used across different platforms.

[0081] In this way, symmetric encryption is used to protect application-layer data, while asymmetric encryption protects session keys, ensuring that even if the transmission packet is intercepted, the data cannot be decrypted. Furthermore, since the base-layer data does not require encryption, it can be shared between different platforms, reducing duplicate transmissions and improving efficiency.

[0082] In step S340, each encrypted transmission package is sent to the corresponding potential cloud cache platform respectively, so that the potential cloud cache platform decrypts the encrypted session key in the encrypted transmission package based on its own cloud cache platform private key to crack the application layer data, thereby restoring the complete task container image based on the cracked application layer data and the basic layer data in the encrypted transmission package.

[0083] In some embodiments, a secure communication channel is established using the TLS / SSL protocol to ensure that the transmission packet cannot be tampered with or stolen during transmission over the network. Preferably, if the transmission path is complex or there are network limitations, the transmission packet can be transmitted in segments over different paths and reassembled on the cloud cache platform.

[0084] Furthermore, on the target cloud cache platform, the encrypted session key in the transmission packet is decrypted using its private key. The decrypted session key is then used to decrypt the encrypted application layer data, retrieving the original application layer content. Based on the base layer data and the decrypted application layer data, the complete task container image is rebuilt on the cloud cache platform. Therefore, each platform can only use its own private key to decrypt the corresponding session key, ensuring that data can only be decrypted and used on the designated platform, thus achieving an isolated data transmission process for different cloud cache platforms.

[0085] In some preferred implementations of the embodiments of the present application, the cloud cache platform public key and the matching cloud cache platform private key for each potential cloud cache platform are a corresponding pair of independent elliptic curve (ECC) public-private key pairs.

[0086] For example, each cloud cache platform (such as A, B, and C) generates its own ECC public and private keys and shares the public keys with the scheduling server in advance. The scheduling server maintains a key library that stores the ECC public keys of each potential cloud cache platform for encrypting the session key.

[0087] It should be noted that compared to the traditional RSA algorithm, ECC has the advantages of shorter key length and higher computational efficiency, making it particularly suitable for resource constraints and efficient scheduling requirements in multi-cloud caching platform environments. Specifically, under the same security conditions, the key length required by ECC is much shorter than that of traditional RSA. For example, the security of ECC-256 bits is equivalent to that of RSA-3072 bits. Therefore, shorter keys mean faster encryption and less memory usage, making it very suitable for multi-cloud caching platform environments with frequent task migration and dynamic scheduling. In addition, ECC has high computational efficiency and low resource overhead, which can speed up encryption and decryption and reduce task startup time.

[0088] Figure 4 A structural connection diagram of an example of a platform scheduling model according to an embodiment of the present application is shown.

[0089] like Figure 4 As shown, the platform scheduling model 400 includes an LSTM layer 410 , a feature concatenation layer 420 , and a simulated annealing matching layer 430 .

[0090] The LSTM layer 410 is used to infer the predicted operating status data of the corresponding cloud cache platform in the next time window based on each historical operating status data.

[0091] Here, LSTM can capture the temporal dependency of historical operating status sequences and predict the resource utilization and latency of each cloud cache platform in the next time window.

[0092] The training set of the LSTM layer 410 can use historical data and generate training samples of different time periods through a sliding window technique. The mean square error (MSE) can be used as a loss function to measure the difference between the predicted operating state and the actual future operating state.

[0093] The feature concatenation layer 420 is used to concatenate the task description data and the predicted running status data to obtain the concatenated feature vectors corresponding to the cloud cache platforms, so as to construct an estimated running status matrix.

[0094] In some implementations, the predicted operating state data (from the LSTM layer) for each cloud cache platform is concatenated with the task description data for the task to be executed to form a concatenated feature vector. The concatenated feature vectors for all cloud cache platforms are then combined to form an estimated operating state matrix, where, for example, each row of the matrix corresponds to a feature vector for a cloud cache platform.

[0095] The simulated annealing matching layer 430 is used to solve the estimated operating state matrix through a simulated annealing algorithm to generate a cross-region data coordination path.

[0096] The simulated annealing (SA) algorithm is a heuristic algorithm that excels at solving complex combinatorial optimization problems. It simulates the heating and cooling of metal annealing to avoid local optima and ultimately find a global optimal solution. Based on this, a set of cloud caching platforms is selected from the estimated operating state matrix and prioritized to generate an optimal cross-regional data collaboration path. This collaborative path provides a basis for dynamic task migration, ensuring that the system can quickly switch to an alternative cloud caching platform if performance bottlenecks are encountered on the first cloud caching platform.

[0097] In some examples of the embodiments of the present application, the LSTM layer 410 is an LSTM layer based on an attention mechanism:

[0098] h i,t ,c i,t =LSTM(x i,t ,h i,t-1 ,c i,t-1 ), Formula (1)

[0099] LSTMs are capable of capturing long- and short-term dependencies in time series data. In cloud computing environments, resource usage across different time periods can exhibit correlations and periodicity. For example, CPU and memory usage fluctuate between peak and off-peak hours, and network latency and bandwidth vary over time. Time series analysis at the LSTM layer enables resource usage prediction for each cloud cache platform, making task scheduling more proactive.

[0100] α i,t =Softmax(W a h i,t ), Formula (2)

[0101] Here, the attention mechanism can reduce the impact of noise or outliers on the model by weighted summation of historical time series states, so that the model can provide reliable predictions even in complex cloud environments (such as random fluctuations in operating status).

[0102]

[0103] It should be noted that the operating status at different time points is not equally important. For example, the resource usage status at certain moments is more valuable for future predictions. The attention mechanism assigns different weights α to different time steps. i,t , ensuring that the model pays more attention to the operating status of the critical time window and improves the prediction accuracy.

[0104]

[0105] Here, the prediction model can output the operating status of each platform at the future time t+1, providing data support for task scheduling decisions. For example, it can predict the CPU usage and latency changes of the AWS platform to help the system determine whether to migrate tasks in advance.

[0106] In the above formula, i represents the i-th cloud cache platform, t represents the t-th time window, and x i,t represents the operating status input vector of the i-th cloud cache platform in the t-th time window, h i,t represents the hidden state of the i-th cloud cache platform in the t-th time window, c i,t represents the LSTM unit state of the i-th cloud cache platform in the t time window; α i,t is the attention weight of the i-th cloud cache platform in the t-th time window, indicating the importance of the moment; W a is the weight matrix of the attention mechanism; N represents the total number of time windows corresponding to the historical running state sequence; is the weighted hidden state of the i-th cloud cache platform, which represents the feature representation after the attention mechanism; It represents the predicted operating status of the i-th cloud cache platform in the t+1-th time window.

[0107] Through the implementation of this application, using an LSTM layer based on an attention mechanism, the scheduling platform can capture both long-term and short-term trends in operational status. This attention mechanism allows the model to focus on time points that have a greater impact on predictions, reducing the interference of noise and abnormal data and improving the robustness of prediction results. This provides accurate prediction data for task migration and allocation, reduces migration losses, and ensures optimal resource utilization and task response efficiency for the cloud cache platform.

[0108] In some examples of the embodiments of the present application, the feature splicing layer 420 is a nonlinear activation layer:

[0109]

[0110] Here, the running status predicted by LSTM Combined with the task description information (data size D, computing resource requirements R, task type L and processing time limit S) into a high-dimensional feature vector Fi Therefore, the demand information of each task is dynamically combined with the operating status information of the cloud cache platform to achieve personalized scheduling for different tasks.

[0111]

[0112] Here, the ReLU activation function and the square term are used to capture the complex nonlinear relationship between task requirements and platform status, and improve the expressiveness of the model. F The influence of different features can be adaptively adjusted, allowing the model to fully utilize key feature information during the scheduling process. In addition, the ReLU activation function can effectively avoid the vanishing gradient problem, ensuring efficient optimization of the model during training.

[0113]

[0114] In this way, the scheduling server can simultaneously analyze the operating status of all cloud cache platforms based on the matrix M and formulate the optimal task scheduling strategy from a global perspective.

[0115] In the above formula, F i represents the concatenated feature vector of the i-th cloud cache platform, D represents the task data size, R represents the task computing resource requirement, L represents the task type, and S represents the task processing time limit; W F and b F Represent the weight matrix and bias term of the nonlinear activation layer respectively, M represents the estimated operating state matrix, represents the feature vector corresponding to the i-th cloud cache platform in M, and n represents the number of cloud cache platforms.

[0116] Through the embodiments of the present application, based on the information in the operating status matrix M, the scheduling server can dynamically assign tasks to ensure that tasks always run on the optimal cloud cache platform. If the operating status of a cloud cache platform changes (such as a predicted surge in CPU or memory usage), the matrix can be used to quickly evaluate the operating status of other platforms and migrate tasks to a more suitable cloud cache platform with minimal migration loss.

[0117] In some examples of the embodiments of the present application, the simulated annealing matching layer 430 is used to perform the following operations:

[0118] The initial path P0 is randomly generated according to the estimated operating state matrix, and the initial path P0 and the initial temperature T are used to calculate the path P0. max Perform annealing iterations.

[0119] In each iteration, the objective function value of the current path P is calculated. The expression of the objective function is:

[0120]

[0121] In an embodiment of the present application, a simulated annealing algorithm is used to find the optimal platform path for the execution of current and future tasks, balancing execution efficiency and migration loss. The first term in the energy function of formula (8) represents the resource consumption of the current task execution, while the second term in formula (8) represents the loss of future task migration. Specifically, through the weighted summation of the feature matrix, the operating status of each platform (such as CPU, memory and network delay) can be dynamically perceived, the most suitable tasks can be assigned, and resource waste can be avoided. In addition, based on the platform characteristics and cost information, the cloud cache platform with lower cost can be given priority to reduce operating expenses.

[0122]

[0123] Here, by using a baseline variable to normalize different metrics, we ensure that all migration loss indicators are compared on the same scale, avoiding errors introduced by differences in data volume, network costs, and latency units. Furthermore, by introducing a squared term for latency variance, we amplify the impact of latency on task performance, ensuring that the system does not easily migrate tasks to high-latency platforms, thereby improving the user experience.

[0124] In the above formula, E(P) represents the energy function value of the current path; x i is a binary variable, indicating whether the task is assigned to platform i; λ is the weight coefficient of migration loss, loss ij represents the loss of migrating tasks from cloud cache platform i to cloud cache platform j; W1, W2 represent the weight matrix used to adjust the relationship between platform features; when a task is migrated from cloud cache platform i to cloud cache platform j, the indicator function otherwise d represents the task data size, d BM represents the amount of benchmark data, c represents the network traffic cost of the cloud cache platform j after migration, and c BM represents the baseline cost, (l j -l i ) represents the latency change from cloud cache platform i to cloud cache platform j, l BM represents the baseline delay; β1, β2, and β3 are weight parameters used to balance the impact of different loss terms.

[0125] Randomly select a new path P′ from the neighborhood of the current path and calculate the acceptance probability of the new path P′:

[0126]

[0127] Where P(accept) represents the probability of acceptance; T g represents the annealing temperature at the g-th iteration, which is used to control the probability of accepting a worse solution.

[0128] Here, during each iteration, the simulated annealing algorithm randomly accepts a suboptimal solution based on the current temperature. This ensures the scheduling system has a certain degree of flexibility and fault tolerance. When faced with sudden changes in operating status, it can explore more scheduling solutions and reduce system bottlenecks caused by resource constraints.

[0129] Perform a cooling operation after each iteration:

[0130] T g+1 =γ·T g , Formula (11)

[0131] Where γ represents the temperature drop coefficient.

[0132] Here, through the above-mentioned cooling operation, it is possible to ensure that the algorithm is fully explored in the early stage and gradually converges to the optimal solution in the later stage, thereby improving the stability of the scheduling system and responding to user requests faster by reducing invalid migration and task blocking.

[0133] When the annealing temperature drops to T min When , the iteration is terminated and the optimal cross-region data collaboration path is output:

[0134] P terget =argmin P {E(P)} , formula (12)

[0135] Where, P terget Indicates cross-region data collaborative path, argmin P {E(P)} represents the path that minimizes the energy function value in multiple iterations.

[0136] In this way, the simulated annealing algorithm avoids the system from falling into local optimality through multiple iterations and neighborhood searches, ensuring that the selected path P target It is the global optimal solution, which dynamically selects the optimal task allocation plan among multiple cloud cache platforms to ensure the best resource utilization and task response time.

[0137] It should be noted that the weights λ and β1, β2, and β3 can be flexibly adjusted according to different task types and business needs. For example, low-latency platforms can be prioritized or platforms with sufficient computing resources can be selected. This enables flexible scheduling and ensures that the system can provide optimized task scheduling solutions in different scenarios.

[0138] The simulated annealing algorithm provided in the embodiments of this application enables global task scheduling to be performed in advance based on prediction results, reducing the number and loss of task migrations, avoiding service interruptions caused by frequent migrations, and improving the reliability of multi-platform intelligent scheduling results. Furthermore, based on the design of the aforementioned objective function or energy function, a comprehensive balance is taken between the resource consumption of the current task execution and the potential loss of future task migrations, and based on the loss function design, an optimal balance can be found between performance and cost, ensuring the efficient operation of the system in a complex multi-platform environment.

[0139] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of combined actions, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application. In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0140] Figure 5 A structural block diagram of an example of a cross-regional data collaborative computing system based on a multi-level cache mechanism according to an embodiment of the present application is shown.

[0141] like Figure 5 As shown, the cross-region data collaborative computing system 500 based on the multi-level cache mechanism includes a data acquisition unit 510, a collaborative path matching unit 520, a task mirror synchronization unit 530 and a platform monitoring switching unit 540.

[0142] The data acquisition unit 510 is used to obtain the historical operating status sequence of each cloud cache platform and the task description data of the target computing task to be collaboratively calculated; the historical operating status sequence includes multiple historical time windows and corresponding historical operating status data, and the parameter types of the historical operating status data include cache hit rate, cache delay, cache capacity utilization, computing resource utilization and network transmission delay; the task description data includes task data size, task data cache requirements, task computing resource requirements, task type and task processing time limit.

[0143] The collaborative path matching unit 520 is used to input the task description data and the historical operating status sequence of each cloud cache platform into the platform scheduling model to output a matching cross-regional data collaborative path; the cross-regional data collaborative path defines multiple potential cloud cache platforms arranged in descending order of priority; the platform scheduling model is a deep learning model.

[0144] The task image synchronization unit 530 is used to package the target computing task and its dependent environment to generate a corresponding task container image, and send the task container image to each of the potential cloud cache platforms.

[0145] The platform monitoring and switching unit 540 is used to control the first potential cache cloud cache platform to run the corresponding task container image, monitor the real-time operating status of the first potential cloud cache platform, and switch to control the second potential cloud cache platform to run the corresponding task container image when the monitored real-time operating status meets the task migration conditions; the priority of the first potential cloud cache platform is higher than that of the second potential cache cloud cache platform.

[0146] In some embodiments, an embodiment of the present application provides a non-volatile computer-readable storage medium, which stores one or more programs including execution instructions, and the execution instructions can be read and executed by an electronic device (including but not limited to a computer, server, or network device, etc.) to execute any of the steps of the above-mentioned cross-regional data collaborative computing method based on a multi-level cache mechanism in the present application.

[0147] In some embodiments, the embodiments of the present application also provide a computer program product, which includes a computer program stored on a non-volatile computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes any step of the above-mentioned cross-regional data collaborative computing method based on a multi-level cache mechanism.

[0148] In some embodiments, an embodiment of the present application also provides an electronic device, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of a cross-regional data collaborative computing method based on a multi-level cache mechanism.

[0149] Figure 6 This is a hardware structure diagram of an electronic device that performs a cross-region data collaborative computing method based on a multi-level cache mechanism, as provided in another embodiment of the present application. Figure 6 As shown, the device includes:

[0150] One or more processors 610 and memory 620, Figure 6 A processor 610 is taken as an example.

[0151] The device for executing the cross-region data collaborative computing method based on the multi-level cache mechanism may further include: an input device 630 and an output device 640 .

[0152] The processor 610, the memory 620, the input device 630 and the output device 640 may be connected via a bus or other means. Figure 6 The bus connection is taken as an example.

[0153] Memory 620, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer executable programs, and modules, such as the program instructions / modules corresponding to the cross-regional data collaborative computing method based on a multi-level cache mechanism in the embodiments of the present application. Processor 610 executes the non-volatile software programs, instructions, and modules stored in memory 620 to execute various functional applications and data processing of the server, thereby implementing the cross-regional data collaborative computing method based on a multi-level cache mechanism in the above-mentioned method embodiment.

[0154] The memory 620 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device, etc. In addition, the memory 620 may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 620 may optionally include a memory remotely located relative to the processor 610, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0155] The input device 630 may receive input digital or character information and generate signals related to user settings and function control of the electronic device. The output device 640 may include a display device such as a display screen.

[0156] The one or more modules are stored in the memory 620 and, when executed by the one or more processors 610, perform the cross-region data collaborative computing method based on the multi-level cache mechanism in any of the above method embodiments.

[0157] The above-mentioned product can execute the method provided in the embodiment of this application, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method provided in the embodiment of this application.

[0158] The electronic devices of the embodiments of the present application exist in various forms, including but not limited to:

[0159] (1) Mobile communication devices: These devices are characterized by their mobile communication capabilities and their primary purpose is to provide voice and data communications. These terminals include smartphones, multimedia phones, feature phones, and low-end phones.

[0160] (2) Ultra-mobile personal computer devices: These devices fall under the category of personal computers and have computing and processing capabilities, and generally also have mobile Internet access. These terminals include PDAs, MIDs, and UMPCs.

[0161] (3) Portable entertainment devices: These devices can display and play multimedia content. They include audio and video players, handheld game consoles, e-books, smart toys, and portable car navigation devices.

[0162] (4) Other onboard electronic devices with data interaction functions, such as onboard computer devices installed in vehicles.

[0163] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0164] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general hardware platform, or of course, by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the relevant technology, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiment.

[0165] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A cross-region data collaborative computing method based on a multi-level cache mechanism, comprising: Obtain the historical operating status sequence of each cloud cache platform and the task description data of the target computing tasks to be collaboratively calculated; The historical operation status sequence includes multiple historical time windows and corresponding historical operation status data. The parameter types of the historical operation status data include cache hit rate, cache delay, cache capacity utilization, computing resource utilization and network transmission delay; the task description data includes task data size, task data cache requirement, task computing resource requirement, task type and task processing time limit; Inputting the task description data and the historical operating status sequence of each of the cloud cache platforms into a platform scheduling model to output a matching cross-regional data collaboration path; The cross-region data collaboration path defines a plurality of potential cloud cache platforms arranged in descending order of priority; The platform scheduling model is a deep learning model; Packaging the target computing task and its dependent environment to generate a corresponding task container image, and sending the task container image to each of the potential cloud cache platforms; Controlling the first potential cloud cache platform to run the corresponding task container image, monitoring the real-time running status of the first potential cloud cache platform, and switching to control the second potential cloud cache platform to run the corresponding task container image if the monitored real-time running status meets the task migration condition; The priority of the first potential cloud cache platform is higher than that of the second potential cloud cache platform; The deep learning model includes an LSTM layer, a feature concatenation layer, and a simulated annealing matching layer; The LSTM layer is used to infer the predicted operating status data of the corresponding cloud cache platform in the next time window based on each of the historical operating status data; The feature splicing layer is used to splice the task description data and each of the predicted operating status data to obtain the corresponding splicing feature vectors of each cloud cache platform to construct an estimated operating status matrix; The simulated annealing matching layer is used to solve the estimated operating state matrix through a simulated annealing algorithm to generate a cross-region data collaborative path.

2. The method according to claim 1, wherein During the process of running the task container image on the first potential cloud cache platform, the first potential cloud cache platform generates a task checkpoint and synchronizes the task checkpoint between each of the potential cloud cache platforms; the task checkpoint is used to store intermediate execution state data of the target computing task; When the monitored first running status data meets the task migration condition, the second potential cloud cache platform is switched to run the corresponding task container image according to the task checkpoint.

3. The method according to claim 1, wherein The LSTM layer is an LSTM layer based on the attention mechanism: ， ， ， ， Where, Indicates the A cloud caching platform, Indicates the t time window, Indicates the The cloud cache platform is t The running state input vector of the time window, Indicates the The cloud cache platform is t The hidden state of the time window, Indicates the Cloud caching platforms t LSTM unit state of time window; For the The cloud cache platform is t The attention weight of the time window; is the weight matrix of the attention mechanism; N Indicates the total number of time windows corresponding to the historical running status sequence; For the The weighted hidden state of a cloud cache platform represents the feature representation after the attention mechanism; Indicates the The cloud cache platform is t + Predicted operating status for 1 time window.

4. The method according to claim 3, wherein: The feature concatenation layer is a nonlinear activation layer: ， ， ， Where, Indicates the The concatenated feature vector of the cloud cache platform, Indicates the task data size, Indicates the task computing resource requirements, Indicates the task type, Indicates the task processing time limit; and denote the weight matrix and bias term of the nonlinear activation layer, respectively. represents the estimated operating state matrix, express Corresponding to The feature vector of a cloud cache platform, n Indicates the number of cloud cache platforms.

5. The method according to claim 4, wherein The simulated annealing matching layer is used to perform the following operations: Randomly generate an initial path according to the estimated operating state matrix , and based on the initial path and initial temperature Perform annealing iteration operation; In each iteration, the current path is calculated The objective function value of ; the expression of the objective function is: ， ， Where, Represents the energy function value of the current path; Is a binary variable indicating whether the task is assigned to the platform ; is the weight coefficient of migration loss, Indicates that the task is from the cloud cache platform Migrate to a cloud caching platform loss; Represents the weight matrix used to adjust the relationship between platform features; When the task is from the cloud cache platform Migrate to a cloud caching platform , indicator function ,otherwise ; Indicates the task data size, Indicates the amount of benchmark data, Represents the cloud cache platform after migration The network traffic cost, represents the baseline cost, Indicates that the cloud cache platform Migrate to a cloud caching platform The delay variation, represents the baseline delay; is a weight parameter used to balance the impact of different loss terms; Randomly select a new path from the neighborhood of the current path , and calculate the new path The acceptance probability is: ， Where, represents the probability of acceptance; Indicates the g The annealing temperature at the iteration is used to control the probability of accepting a worse solution; Perform a cooling operation after each iteration: ， Where, represents the temperature reduction coefficient; When the annealing temperature drops to When , the iteration is terminated and the optimal cross-region data collaboration path is output: ， Where, Indicates the cross-region data collaboration path, Represents the path that minimizes the energy function value in multiple iterations.

6. The method according to claim 1, wherein The step of packaging the target computing task and its dependent environment to generate a corresponding task container image, and sending the task container image to each of the potential cloud cache platforms, includes: Analyze the task code, dependency packages and environment configuration corresponding to the target computing task; Call the container image tool to package the parsed task code, dependency packages, and environment configuration to generate the corresponding initial image file; The initial image file is encrypted to obtain a corresponding task container image, and the task container image is sent to each of the potential cloud cache platforms.

7. The method according to claim 6, wherein: The encrypting the initial image file to obtain a corresponding task container image, and sending the task container image to each of the potential cloud cache platforms, includes: Parsing the base layer data and application layer data in the initial image file, and obtaining a random session key and a cloud cache platform public key for each potential cloud cache platform; symmetrically encrypting the application layer data according to each of the random session keys to generate corresponding encrypted application layer data; For each potential cloud cache platform, asymmetrically encrypt the corresponding random session key using the cloud cache platform public key of the potential cloud cache platform to obtain a corresponding encrypted session key, and package the basic layer data, the encrypted application layer data, and the encrypted session key into a corresponding encrypted transmission package; Each of the encrypted transmission packages is sent to the corresponding potential cloud cache platform respectively, so that the potential cloud cache platform decrypts the encrypted session key in the encrypted transmission package based on its own cloud cache platform private key to crack the application layer data, thereby restoring the complete task container image based on the cracked application layer data and the base layer data in the encrypted transmission package.

8. The method according to claim 7, wherein: The cloud cache platform public key and the corresponding cloud cache platform private key for each potential cloud cache platform are a corresponding pair of independent elliptic curve public-private key pairs.

9. A cross-regional data collaborative computing system based on a multi-level cache mechanism, comprising: A data acquisition unit is used to obtain the historical operation status sequence of each cloud cache platform and the task description data of the target computing task to be collaboratively calculated; The historical operation status sequence includes multiple historical time windows and corresponding historical operation status data. The parameter types of the historical operation status data include cache hit rate, cache delay, cache capacity utilization, computing resource utilization and network transmission delay; the task description data includes task data size, task data cache requirement, task computing resource requirement, task type and task processing time limit; A collaborative path matching unit, configured to input the task description data and the historical operating status sequence of each of the cloud cache platforms into a platform scheduling model to output a matching cross-regional data collaborative path; The cross-region data collaboration path defines a plurality of potential cloud cache platforms arranged in descending order of priority; The platform scheduling model is a deep learning model; A task image synchronization unit, configured to package the target computing task and its dependent environment to generate a corresponding task container image, and send the task container image to each of the potential cloud cache platforms; a platform monitoring and switching unit, configured to control the first potential cloud cache platform to run the corresponding task container image, monitor the real-time running status of the first potential cloud cache platform, and switch to control the second potential cloud cache platform to run the corresponding task container image if the monitored real-time running status meets the task migration condition; The priority of the first potential cloud cache platform is higher than that of the second potential cloud cache platform; The deep learning model includes an LSTM layer, a feature concatenation layer, and a simulated annealing matching layer; The LSTM layer is used to infer the predicted operating status data of the corresponding cloud cache platform in the next time window based on each of the historical operating status data; The feature splicing layer is used to splice the task description data and each of the predicted operating status data to obtain the corresponding splicing feature vectors of each cloud cache platform to construct an estimated operating status matrix; The simulated annealing matching layer is used to solve the estimated operating state matrix through a simulated annealing algorithm to generate a cross-region data collaborative path.

Images