Cross-cloud container arrangement method and system based on binary labels

Abstract

The invention discloses a cross-cloud container arrangement method and system based on a binary tag, and the method comprises the steps: an operation cloud generates a control frame containing an authentication key, screens a target container tag, and broadcasts the target container tag to an assistance cloud; after assisting cloud verification signature, feeding back a public container label and interface metadata; the operation cloud constructs a cross-cloud arrangement strategy, packages the strategy into an update management and control frame and pushes the update management and control frame to a managed server; the managed server dynamically calculates the weight of the container instance and updates the forwarding table; the transmission management server generates a label extension head and intelligently adapts to a target cloud protocol, the extension head is added when a label protocol is started, and a data packet is reconstructed according to an API specification when the label protocol is not started; and the checking module monitors the execution deviation in real time and triggers a strategy to dynamically adjust a closed loop. Containers are accurately screened through binary labels, and efficiency can be effectively improved; the transmission delay of the heterogeneous cloud platform is reduced through an intelligent data encapsulation mechanism; the strategy adjustment response time is shortened based on deviation feedback of real-time monitoring; and the cross-cloud authentication success rate is improved through bidirectional signature verification.

Description

Technical Field

[0001] The present invention relates to the fields of cloud computing and artificial intelligence, and specifically to a cross-cloud container orchestration method and system based on binary tags. Background Art

[0002] Currently, with the continuous popularization and application of cloud computing and middleware technologies, related container technologies and container orchestration technologies have also been widely used. Among them, although container orchestration technology emerged relatively late, it can effectively organize functional containers to form a service group cluster and achieve the functions of high concurrency and high availability of the cloud platform, and is currently attracting much attention. However, there are still certain problems with existing container orchestration technologies: First, existing orchestration technologies are based on foreign Kubernetes container management technologies, and once restricted, it will seriously affect the overall security of domestic clouds; second, existing orchestration technologies can well orchestrate container resources in the same cloud environment or two cloud environments with the same architecture, but cannot achieve container orchestration in more than three multi-cloud and different architecture cloud environments. Third, existing technologies pay more attention to the organization and implementation of business functions, and all container orchestration is completed manually, unable to achieve program automation orchestration, nor can it optimize the automation orchestration results. Summary of the Invention

[0003] The present invention provides a cross-cloud container orchestration method and system based on binary tags, which can achieve automated container orchestration in multi-cloud and different architecture cloud environments.

[0004] A cross-cloud container orchestration method based on binary tags includes the following steps:

[0005] Step S1: The management server of the operating cloud receives a cross-cloud orchestration request initiated by the operating cloud, generates a control frame containing an authentication key and container label screening conditions, and sends the control frame to at least one assisting cloud;

[0006] Step S2: The management server of the assisting cloud verifies the signature legality of the control frame. If the verification passes, it feeds back a list of publicly available container labels and corresponding interface metadata to the operating cloud. The interface metadata includes input data format, output status code, and container resource occupancy rate;

[0007] Step S3: The management server of the operating cloud generates a cross-cloud orchestration strategy according to the container label list and interface metadata. The strategy includes:

[0008] Container group sequence: A call path defined by a primary-assistant mode, a chained mode, or a hybrid mode based on binary tags;

[0009] Routing weight: A data forwarding priority dynamically allocated according to the container resource occupancy rate;

[0010] Step S4: The management server of the operation cloud encapsulates the cross-cloud orchestration policy into an update control frame and pushes it to the managed servers and the transmission management server of the operation cloud and the assisting cloud;

[0011] Step S5: The managed server recalculates the container instance weights according to the container group sequence and the container instance status information in the update control frame, and updates the local forwarding table; the transmission management server generates a label extension header for the cross-cloud data frame according to the update control frame, and the extension header includes a cross-cloud identifier, a container group sequence, and an encryption signature;

[0012] Step S6: When the transmission management server performs cross-cloud data transmission, if the peer cloud enables the label protocol, it adds the label extension header to the data frame; if the peer cloud does not enable the protocol, it repackages the data payload according to the target cloud API specification;

[0013] Step S7: The verification module monitors the output results of the container instances. If the deviation between the actual result and the expected result exceeds the threshold, it sends a policy adjustment signal to the management server;

[0014] Step S8: The management server re-executes Steps S3 - S6 according to the policy adjustment signal, generates an updated cross-cloud orchestration policy, and distributes it.

[0015] Further, the generation of the cross-cloud orchestration policy in Step S3 specifically includes:

[0016] Primary-assistant mode: Select the primary container label from the container label list and bind at least one assisting container label to form a "primary container → assisting container" call chain;

[0017] Chain mode: Generate a pipeline call path according to the binary sequence order of the container labels;

[0018] Hybrid mode: Insert a branch marker (0xFF00) into the call chain to switch the logic between the primary-assistant mode and the chain mode.

[0019] Further, the update of the forwarding table of the managed server in Step S5 includes:

[0020] Establish a mapping relationship between the label and the container instance IP address according to the container group sequence;

[0021] Allocate a load balancing ratio for multiple instances of the same container group according to the routing weight;

[0022] If a container instance failure is detected, send an instance failure notice to the management server and the managed server in the same domain.

[0023] Further, the execution logic of the verification module in Step S7 includes:

[0024] Supervised learning mechanism: Compare the actual output status code with the expected status code, calculate the error rate, and generate a feedback signal;

[0025] Unsupervised learning mechanism: Evaluate the score of the orchestration strategy according to predefined rules, and select the strategy with the highest score as the optimization target. The predefined rules include minimizing latency and maximizing the fault tolerance rate.

[0026] Furthermore, the encapsulation rule of the cross-cloud data frame in step S6 is:

[0027] If the peer cloud enables the label protocol, the data frame structure includes:

[0028] Standard header: source cloud ID, target cloud ID, protocol version;

[0029] Label extension header: container group sequence, encryption signature, routing weight;

[0030] Data payload: native service data;

[0031] If the peer cloud does not enable the protocol, the data frame structure is converted to the target cloud API format, including:

[0032] RESTful request: Map the container group sequence to the URL path parameter;

[0033] gRPC call: Encode the content of the label extension header as Protobuf metadata.

[0034] A cross-cloud container orchestration device based on binary labels, including:

[0035] Management server module, used for:

[0036] Receive a cross-cloud orchestration request initiated by the operation cloud, generate and send a control frame containing an authentication key and container label filtering conditions to the assisting cloud;

[0037] Parse the container label list and interface metadata returned by the assisting cloud, and generate a cross-cloud orchestration strategy based on the label binary sequence. The strategy includes a container group sequence, a routing weight, and an operating mode;

[0038] Encapsulate the orchestration strategy into an updated control frame and push it to the managed server module and the transmission management server module;

[0039] Receive a strategy adjustment signal sent by the verification module, regenerate the orchestration strategy, and iterate and issue it;

[0040] Managed server module, used for:

[0041] Extract the container group sequence and container instance status information from the updated control frame, calculate the weight, update the local forwarding table, and establish a mapping relationship between the label and the container instance and a load balancing rule;

[0042] Monitor the status of local container instances in real time and send status change information to the management server and other managed servers in a timely manner;

[0043] Execute data forwarding according to the forwarding table. If the running mode is the master-slave mode, call the assisting container preferentially according to the master container label; if it is the chained mode, trigger pipeline processing in the order of the label sequence;

[0044] The transmission management server module is used for:

[0045] Extract the container group sequence and container instance status information from the update control frame, calculate the weight, and generate a label extension header, which includes a cross-cloud identifier, an encryption signature, and a container group call path;

[0046] When performing cross-cloud data transmission, if the peer enables the label protocol, add the label extension header to the data frame; if the protocol is not enabled, convert and encapsulate the data payload according to the target cloud API specification;

[0047] The verification module is used for:

[0048] Monitor the actual output result of the container instance, compare it with the expected result and calculate the deviation value, and generate a policy adjustment signal when the deviation exceeds the threshold;

[0049] Send the policy adjustment signal to the management server module to trigger the dynamic iteration of the cross-cloud orchestration policy.

[0050] A cross-cloud container orchestration system based on binary labels, including: a computer-readable storage medium and a processor;

[0051] The computer-readable storage medium is used to store executable instructions;

[0052] The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the cross-cloud container orchestration method based on binary labels.

[0053] A non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the cross-cloud container orchestration method based on binary labels.

[0054] The present invention precisely screens containers through binary tags, realizes data allocation and forwarding during network transmission, and can effectively improve transmission efficiency; the use of binary tags facilitates the processing of dedicated or general-purpose processing chips, further accelerating the transmission efficiency; the intelligent data encapsulation mechanism enables interoperability between heterogeneous cloud platforms, and cloud-to-cloud communication no longer depends on products of specific manufacturers; the support for automated program orchestration improves container orchestration efficiency and expands the application scenarios of container orchestration; the deviation feedback shortening strategy based on real-time monitoring adjusts the response time and improves the efficiency of automated orchestration. Description of the Drawings

[0055] Figure 1 is a schematic flowchart of a cross-cloud container orchestration method based on binary tags according to an embodiment of the present invention;

[0056] Figure 2 is an orchestration schematic diagram of enabling a label forwarding system according to an embodiment of the present invention;

[0057] Figure 3 is a schematic diagram of the running mode of a container instance - master-slave mode according to an embodiment of the present invention;

[0058] Figure 4 is a schematic diagram of the running mode of a container instance - chain mode according to an embodiment of the present invention;

[0059] Figure 5 is a schematic diagram of the running mode of a container instance - hybrid mode according to an embodiment of the present invention;

[0060] Figure 6 is a schematic flowchart of operating a cloud according to an embodiment of the present invention;

[0061] Figure 7 is a schematic flowchart of assisting a cloud according to an embodiment of the present invention;

[0062] Figure 8 is a schematic diagram of cloud-to-cloud orchestration without enabling a label protocol according to an embodiment of the present invention;

[0063] Figure 9 is a schematic diagram of container and functional automated orchestration according to an embodiment of the present invention. Detailed Embodiments

[0064] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0065] The technical terms used in the embodiments of the present invention are introduced as follows:

[0066] The management server is a collective term for one or a group of computing resources that can manage cloud containers and business functions.

[0067] The managed server is a collective term for one or a group of computing resources that actually run containers to provide computing capabilities externally and forward input / output data frame tags.

[0068] The transmission management server realizes cross-cloud container data transmission, especially functions such as interface calls between heterogeneous clouds and data repackaging.

[0069] The orchestratable content is the container orchestration result or container and function information publicly disclosed by the relevant cloud and can be used by other clouds.

[0070] In addition to storing and managing container tags, the management server also stores the external service interface information of the container, including but not limited to the input data format and content, and the output data format and content in normal or abnormal situations, for use in container orchestration and construction functions. Container orchestration is performed by manual or automated programs and is completed independently by the local cloud management server or through negotiation among multiple cloud management servers. The orchestration result is stored and managed by the management server.

[0071] If the container orchestration result changes, the management server notifies the managed server and the transmission management server. The managed server and the transmission management server query their own forwarding tables, and after confirming the need for update, send update requests to the management server. After receiving the request, the management server pushes the updated content such as the updated orchestration result sequence and routing weight to the two groups of servers. The managed server updates the local forwarding policy according to the received updated content, and the transmission management server updates the transmission forwarding policies such as the data packaging frame method and the heterogeneous cloud call method according to the received updated content.

[0072] In the label forwarding system, these updates are realized by the transmission of control frames during the service period.

[0073] Please refer to Figures 1-9 , the embodiments of the present invention provide a cross-cloud container orchestration method based on binary tags, including the following steps:

[0074] Step S1: The management server of the operating cloud receives a cross-cloud orchestration request initiated by the operating cloud, generates a control frame containing an authentication key and a container tag screening condition, and sends the control frame to at least one assisting cloud (as Figure 6 shown);

[0075] Step S2: Assist the management server of the cloud to verify the signature legality of the control frame. If the verification passes, feedback the list of publicly available container tags and the corresponding interface metadata to the operating cloud. The interface metadata includes the input data format, output status code, and container resource occupancy rate;

[0076] Step S3: The management server of the operating cloud generates a cross-cloud orchestration policy based on the container tag list and interface metadata. The cross-cloud orchestration policy includes:

[0077] Container group sequence: The main-assist mode, chain mode, or hybrid mode call path defined based on binary tags;

[0078] Routing weight: The data forwarding priority dynamically allocated according to the container resource occupancy rate;

[0079] Among them, the generation of the cross-cloud orchestration policy specifically includes:

[0080] Main-assist mode (as Figure 3 shown): Select the main container tag from the container tag list and bind at least one assist container tag to form a "main container → assist container" call chain;

[0081] Chain mode (as Figure 4 shown): Generate a pipeline call path according to the binary sequence order of the container tags;

[0082] Hybrid mode (as Figure 5 shown): Insert a branch mark (0xFF00) in the call chain to switch the main-assist mode and chain mode logic.

[0083] Step S4: The management server of the operating cloud encapsulates the cross-cloud orchestration policy into an updated control frame and pushes it to the managed servers and transmission management servers of the operating cloud and the assist cloud (as Figure 6 shown);

[0084] Step S5: The managed server recalculates the container instance weight according to the container group sequence and container instance status information in the updated control frame and updates the local forwarding table; The transmission management server generates a label extension header for the cross-cloud data frame according to the updated control frame, and the extension header includes a cross-cloud identifier, a container group sequence, and an encrypted signature;

[0085] Among them, the update of the forwarding table of the managed server includes:

[0086] Establish a mapping relationship between the tag and the container instance IP address according to the container group sequence;

[0087] Allocate the load balancing ratio for multiple instances of the same container group according to the routing weight;

[0088] If a container instance failure is detected, send an instance failure notice to the management server and the managed servers in the same domain.

[0089] Step S6: When the transfer management server performs cross-cloud data transfer, if the peer cloud enables the label protocol, add the label extension header to the data frame (as Figure 8 shown); if the peer cloud does not enable the protocol, repackage the data payload according to the target cloud API specification;

[0090] Among them, the encapsulation rules for cross-cloud data frames are:

[0091] If the peer cloud enables the label protocol, the data frame structure includes:

[0092] Standard header: source cloud ID, target cloud ID, protocol version;

[0093] Label extension header: container group sequence, encryption signature, routing weight;

[0094] Data payload: native service data;

[0095] If the peer cloud does not enable the protocol, the data frame structure is converted to the target cloud API format, including:

[0096] RESTful request: Map the container group sequence to the URL path parameter;

[0097] gRPC call: Encode the label extension header content as Protobuf metadata.

[0098] Step S7: The verification module monitors the output result of the container instance. If the deviation between the actual result and the expected result exceeds the threshold, send a policy adjustment signal to the management server;

[0099] Among them, the execution logic of the verification module includes:

[0100] Supervised learning mechanism: Compare the actual output status code with the expected status code, calculate the error rate and generate a feedback signal;

[0101] Unsupervised learning mechanism: Evaluate the orchestration policy score according to predefined rules (such as minimizing latency and maximizing fault tolerance), and select the policy with the highest score as the optimization target.

[0102] When the embodiment of the present invention applies automated orchestration to cross-cloud orchestration, the operating cloud can obtain information such as assisted cloud container information, container call information, container orchestration results, or status information of cloud function components marked by tags. Then, according to predefined automated orchestration rules or by verifying the data processing results of the container instance group after orchestration, the container orchestration is completed automatically. The automated orchestration according to predefined rules is similar to unsupervised learning in machine learning, and the optimal orchestration result among numerous automated orchestrations is found according to the rules; the orchestration by verifying the data processing results is similar to supervised learning in machine learning. By setting a verification module (such as Figure 9 as shown) after a whole or a certain section of the orchestrated process, whether the output result after orchestration meets the requirements is verified manually or by machine to determine the best orchestration result.

[0103] The automated orchestration can also be combined with the AI automatic programming function. On the one hand, it continuously optimizes the programs running in the containers, and on the other hand, it continuously adjusts the entire orchestration process to ensure that the orchestration process meets the needs.

[0104] Since the orchestration result can be applied to multiple groups of container instances, parallel operation of multiple groups of container instances can be achieved. Combining with the comprehensive data processing ability of the verification program for multiple groups of data results, more advanced functions can also be realized.

[0105] The container instances after automated orchestration can also run in the master-slave mode, chain mode, or hybrid mode locally or across clouds.

[0106] Step S8: The management server adjusts the signal according to the policy, re-executes steps S3 - S6, generates an updated cross-cloud orchestration policy, and distributes it.

[0107] The embodiment of the present invention also provides a cross-cloud container orchestration device based on binary tags for executing the above method, including:

[0108] The management server module is used for:

[0109] Receiving a cross-cloud orchestration request initiated by the operating cloud, generating and sending a control frame containing an authentication key and container label screening conditions to the assisted cloud;

[0110] Parsing the container label list and interface metadata returned by the assisted cloud, generating a cross-cloud orchestration policy based on the label binary sequence, and the policy includes a container group sequence, routing weights, and an operating mode;

[0111] Encapsulating the orchestration policy into an updated control frame and pushing it to the managed server module and the transmission management server module;

[0112] Receiving a policy adjustment signal sent by the verification module, regenerating the orchestration policy, and iteratively distributing it;

[0113] The managed server module is used for:

[0114] Extract the container group sequence and container instance status information from the update control frame, calculate the weights, and update the local forwarding table, and establish the mapping relationship between the labels and the container instances and the load balancing rules;

[0115] Monitor the status of local container instances in real time, and timely send status change information to the management server and other managed servers;

[0116] Execute data forwarding according to the forwarding table. If the running mode is the master-assistant mode, call the assisting containers preferentially according to the master container label; if it is the chain mode, trigger the pipeline processing in the order of the label sequence;

[0117] The transmission management server module is used for:

[0118] Extract the container group sequence and container instance status information from the update control frame, calculate the weights, and generate a label extension header, and the extension header includes the cross-cloud identifier, the encryption signature, and the container group call path;

[0119] When performing cross-cloud data transmission, if the peer enables the label protocol, add the label extension header to the data frame; if the protocol is not enabled, convert and encapsulate the data payload according to the target cloud API specification;

[0120] The verification module is used for:

[0121] Monitor the actual output results of the container instances, compare with the expected results and calculate the deviation value, and generate a policy adjustment signal when the deviation exceeds the threshold;

[0122] Send the policy adjustment signal to the management server module to trigger the dynamic iteration of the cross-cloud orchestration policy.

[0123] Another embodiment of the present invention provides a cross-cloud container orchestration system based on binary labels, including: a computer-readable storage medium and a processor;

[0124] The computer-readable storage medium is used to store executable instructions;

[0125] The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the cross-cloud container orchestration method based on binary labels.

[0126] Another embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the cross-cloud container orchestration method based on binary labels is implemented.

[0127] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0128] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0129] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0130] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Therefore, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: still modifications or equivalent replacements can be made to the specific embodiments of the present invention, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A cross-cloud container orchestration method based on binary tags, characterized in that, It includes the following steps: Step S1: The management server of the operating cloud receives a cross-cloud orchestration request initiated by the operating cloud, generates a control frame containing an authentication key and container label filtering conditions, and sends the control frame to at least one assisting cloud; Step S2: The management server of the assisting cloud verifies the signature legality of the control frame. If the verification passes, it feeds back a list of publicly available container labels and corresponding interface metadata to the operating cloud. The interface metadata includes input data format, output status code, and container resource occupancy rate; Step S3: The management server of the operating cloud generates a cross-cloud orchestration policy based on the container label list and interface metadata. The policy includes: Container group sequence: A primary-assistant mode, chain mode, or hybrid mode call path defined based on binary labels; Routing weight: A data forwarding priority dynamically allocated according to the container resource occupancy rate; Step S4: The management server of the operating cloud encapsulates the cross-cloud orchestration policy into an updated control frame and pushes it to the managed servers and transmission management servers of the operating cloud and the assisting cloud; Step S5: The managed server recalculates the container instance weight based on the container group sequence and container instance status information in the updated control frame and updates the local forwarding table; The transmission management server generates a label extension header for the cross-cloud data frame according to the updated control frame. The extension header includes a cross-cloud identifier, a container group sequence, and an encrypted signature; Step S6: When the transmission management server performs cross-cloud data transmission, if the peer cloud enables the label protocol, it adds the label extension header to the data frame; if the peer cloud does not enable the protocol, it repackages the data payload according to the target cloud API specification; Step S7: The verification module monitors the output results of the container instances. If the deviation between the actual result and the expected result exceeds the threshold, it sends a policy adjustment signal to the management server; Step S8: The management server, according to the policy adjustment signal, re-executes steps S3 - S6, generates an updated cross-cloud orchestration policy, and distributes it; 2. The method according to claim 1, wherein The generation of the cross-cloud orchestration policy in step S3 specifically includes: Primary-assistant mode: Select a primary container label from the container label list and bind at least one assisting container label to form a "primary container → assisting container" call chain; Chain mode: Generate a pipeline call path in the order of the binary sequence of the container labels; Hybrid mode: Insert a branch marker (0xFF00) in the call chain to switch the logic between the primary-assistant mode and the chain mode; 3. The method according to claim 1, wherein The update of the forwarding table of the managed server in step S5 includes: Establish a mapping relationship between the label and the container instance IP address according to the container group sequence; Allocate a load balancing ratio for multiple instances of the same container group according to the routing weight; If a container instance failure is detected, send an instance failure notification to the management server and the managed server in the same domain; 4. The method according to claim 1, wherein The execution logic of the verification module in step S7 includes: Supervised learning mechanism: Compare the actual output status code with the expected status code, calculate the error rate, and generate a feedback signal; Unsupervised learning mechanism: Evaluate the orchestration policy score according to predefined rules, and select the policy with the highest score as the optimization target. The predefined rules include minimizing latency and maximizing fault tolerance; 5. The method according to claim 1, wherein The encapsulation rule of the cross-cloud data frame in step S6 is: If the peer cloud enables the label protocol, the data frame structure includes: Standard header: source cloud ID, destination cloud ID, protocol version; Label extension header: container group sequence, encryption signature, routing weight; Data payload: native service data; If the peer cloud does not enable the protocol, the data frame structure is converted to the target cloud API format, including: RESTful request: map the container group sequence to URL path parameters; gRPC call: encode the content of the label extension header as Protobuf metadata.

6. A cross-cloud container orchestration device based on binary tags, characterized in that Including: Management server module, used for: Receiving a cross-cloud orchestration request initiated by the operation cloud, generating and sending a control frame containing an authentication key and container label filtering conditions to the assisting cloud; Parsing the container label list and interface metadata returned by the assisting cloud, generating a cross-cloud orchestration policy based on the label binary sequence, the policy including container group sequence, routing weight, and operating mode; Encapsulating the orchestration policy into an updated control frame and pushing it to the managed server module and the transmission management server module; Receiving a policy adjustment signal sent by the verification module, regenerating the orchestration policy and iteratively distributing it; Managed server module, used for: Extracting the container group sequence and container instance status information from the updated control frame, calculating weights and updating the local forwarding table, establishing a mapping relationship between labels and container instances, and load balancing rules; Real-time monitoring of the local container instance status, and timely sending status change information to the management server and other managed servers; Performing data forwarding according to the forwarding table. If the operating mode is the primary cooperation mode, call the assisting container preferentially according to the primary container label; if it is the chained mode, trigger pipeline processing in the order of the label sequence; Transmission management server module, used for: Extracting the container group sequence and container instance status information from the updated control frame, calculating weights, generating a label extension header, the extension header including cross-cloud identification, encryption signature, and container group call path; During cross-cloud data transmission, if the peer enables the label protocol, add the label extension header to the data frame; If the protocol is not enabled, convert and encapsulate the data payload according to the target cloud API specification; Verification module, used for: Monitoring the actual output result of the container instance, comparing it with the expected result and calculating the deviation value, and generating a policy adjustment signal when the deviation exceeds the threshold; Sending the policy adjustment signal to the management server module to trigger the dynamic iteration of the cross-cloud orchestration policy.

7. A cross-cloud container orchestration system based on binary tags, comprising: Computer-readable storage medium and processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the cross-cloud container orchestration method based on binary labels according to any one of claims 1-5.

8. A non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the cross-cloud container orchestration method based on binary labels according to any one of claims 1-5.

Images