Sandbox platform technical method and system of Kubernetes and OpenStack shared node based on dynamic memory resource scheduling

Abstract

The invention discloses a dynamic memory resource scheduling-based sandbox platform technical method and system for sharing nodes by Kubernetes and OpenStack, and belongs to the technical field of cloud computing, and the method comprises the following steps of: acquiring a global memory state from an OVN controller by developing an OVN scheduling module, cooperatively managing memory allocation of an OpenStack sandbox and the Kubernetes sandbox based on a unified strategy, and managing the memory allocation of the OpenStack sandbox and the Kubernetes sandbox. Comprising a memory scheduling process combined with sandbox business logic and a memory scheduling process combined with K8s business logic, so that accurate scheduling of dynamic memory resources is realized, and the memory use efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of cloud computing technology, specifically to a sandbox platform technology method and system for Kubernetes and OpenStack shared nodes based on dynamic memory resource scheduling. Background Technology

[0002] In the development of cloud computing, in order to improve resource utilization and reduce costs, deploying OpenStack and Kubernetes (K8s) on shared nodes has become a trend.

[0003] However, because OpenStack allocates memory based on sandbox specifications, while Kubernetes allocates memory based on the resource requests and limits of the sandbox, the two have different memory scheduling strategies. In addition, the dynamic changes in business load lead to many challenges in memory scheduling on shared nodes, such as system performance degradation and stability issues caused by memory over-provisioning.

[0004] Meanwhile, in real-world business scenarios, a secure, isolated sandbox environment that can flexibly utilize resources is required, and existing technologies are insufficient to meet the need to build an efficient sandbox platform in such a complex environment. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a sandbox platform technology method and system for Kubernetes and OpenStack shared nodes based on dynamic memory resource scheduling, so as to solve the problems mentioned in the background art.

[0006] The objective of this invention can be achieved through the following technical solution: a sandbox platform technology method for Kubernetes and OpenStack shared nodes based on dynamic memory resource scheduling, comprising the following steps: Global memory status information is obtained from the OVN controller. Based on a unified scheduling strategy, memory allocation for OpenStack sandboxes and Kubernetes sandboxes is managed collaboratively. This includes memory scheduling processes that combine sandbox business logic and memory scheduling processes that combine Kubernetes business logic. The global memory status information includes the memory usage of nodes in each type of sandbox. When the sandbox creation process starts, the current node's memory status information is obtained synchronously and recorded in the OVN controller; Based on node memory usage and sandbox memory requirements, a preliminary memory allocation plan is made; memory scheduling strategies are adjusted by updating node memory resource status, tracking memory release, and coordinating memory resource allocation. Based on the latest remaining resource information, and considering the sandbox and global memory requirements, memory allocation is replanned.

[0007] Preferably, the memory scheduling process combined with sandbox business logic includes: The system initiates the sandbox creation process, and the real-time monitoring module synchronously obtains the current node's memory status information and records it to the OVN controller. Select and confirm the use of node1. The OVN controller will make an initial memory allocation plan based on the current memory usage of node1 and the sandbox memory requirements. Nova sends a node occupancy notification to the OVN controller. The OVN controller updates the memory resource status of node1 and adjusts the memory scheduling policy according to the new status. At the same time, the OVN controller detects the change that memory resources are about to be occupied.

[0008] Preferably, the memory scheduling process combined with sandbox business logic further includes: The kubelet is taken offline via the OVN controller. During this process, the OVN controller tracks memory release and provides feedback to the OVN controller, which then adjusts memory allocation based on the memory changes. After Nova completes virtual machine creation, it obtains the remaining resource information and updates it to the OVN controller. The intelligent allocation module then uses the remaining resource information, combined with the sandbox and global memory requirements, to re-plan memory allocation. The kubelet is brought online through the OVN controller to obtain the latest remaining resource information and updates the OVN controller with the latest remaining resource information; the OVN controller further coordinates memory resource allocation based on the latest remaining resource information.

[0009] Preferably, the memory scheduling process combined with Kubernetes business logic includes: When K8s starts the sandbox creation process, the OVN controller obtains the current node's memory state information and records it to the OVN controller. Select and confirm the use of node1. The OVN controller will make an initial memory allocation plan based on the current memory usage of node1 and the sandbox memory requirements. The kubelet sends a node occupancy notification to the OVN controller, which then updates the memory resource status of node1 and adjusts the memory scheduling policy accordingly. Simultaneously, the OVN controller monitors changes in memory resource occupancy.

[0010] Preferably, the memory scheduling process combined with Kubernetes business logic further includes: The OVN controller performs the offline nova operation, tracks the memory release status, and feeds back to the intelligent allocation module. The OVN controller further adjusts the memory allocation based on the memory changes. After the kubelet completes the sandbox creation, it obtains the remaining resource information and updates the remaining resource information to the OVN controller; the OVN controller, based on the remaining resource information and the sandbox and global memory requirements, re-plans the memory allocation. The kubelet is brought online through the OVN controller to obtain the latest remaining resource information and updates the OVN controller with the latest remaining resource information; the OVN controller then coordinates memory resource allocation again based on the latest remaining resource information.

[0011] To address the aforementioned issues, this invention also provides a sandbox platform technology system for Kubernetes and OpenStack shared nodes based on dynamic memory resource scheduling, comprising: The OVN scheduling module is used to obtain global memory status information from the OVN controller and to coordinate the memory allocation of OpenStack sandbox and Kubernetes sandbox based on a unified scheduling strategy. The real-time monitoring module is used to synchronously acquire the current node's memory status information when the sandbox creation process starts and record it to the OVN controller; The OVN controller is used to initially plan memory allocation based on node memory usage and sandbox memory requirements; adjust memory scheduling strategies by updating node memory resource status, track memory release, and coordinate memory resource allocation; The intelligent allocation module is used to re-plan memory allocation based on the latest remaining resource information, combined with sandbox and global memory requirements.

[0012] Preferably, nova is used to send node occupancy notifications to the OVN controller, create virtual machines, and update the remaining resource information to the OVN controller.

[0013] Preferably, kubelet is used to send node occupancy notifications to the OVN controller, create sandboxes, and update the remaining resource information to the OVN controller.

[0014] Compared to existing solutions, the beneficial effects achieved by this invention are: This invention develops an OVN scheduling module to achieve collaborative management of memory allocation in OpenStack and K8s sandboxes, enabling precise allocation of memory resources and avoiding waste and unreasonable use of resources. Among them, the memory scheduling process that combines sandbox business logic and Kubernetes business logic effectively solves the problem of memory conflicts on shared nodes of Kubernetes and OpenStack in the sandbox platform by managing memory in the stages of sandbox creation, node occupancy notification, resource creation and update, and resource deployment, thus ensuring business continuity and enhancing system stability. This invention obtains node memory status information through a real-time monitoring module, providing data support for memory scheduling; the intelligent allocation module re-plans memory allocation based on remaining resource information, dynamically adapting to changes in business load, adjusting the memory allocation plan in a timely manner according to actual needs, improving memory utilization efficiency, and enhancing overall system performance. Attached Figure Description

[0015] The invention will now be further described with reference to the accompanying drawings.

[0016] Figure 1 This is a flowchart of the memory scheduling process that incorporates sandbox business logic in this invention.

[0017] Figure 2 This is a flowchart of the memory scheduling process that incorporates K8s business logic in this invention.

[0018] Figure 3 This is a module structure diagram of a sandbox platform technology system for Kubernetes and OpenStack shared nodes based on dynamic memory resource scheduling, as proposed in this invention. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0020] Example 1, such as Figure 1 and Figure 2 As shown, this invention is a sandbox platform technology method for Kubernetes and OpenStack shared nodes based on dynamic memory resource scheduling, including the following steps: Obtain global memory status information from the OVN controller, and coordinate the memory allocation of OpenStack sandbox and Kubernetes (K8s) sandbox based on a unified scheduling strategy. This includes a memory scheduling process that combines sandbox business logic and a memory scheduling process that combines K8s business logic. When the sandbox creation process starts, the current node's memory status information is obtained synchronously and recorded in the OVN controller; Based on node memory usage and sandbox memory requirements, a preliminary memory allocation plan is made; memory scheduling strategies are adjusted by updating node memory resource status, tracking memory release, and coordinating memory resource allocation. Based on the latest remaining resource information, and considering the sandbox and global memory requirements, memory allocation is replanned.

[0021] In some preferred implementations, data interaction is conducted through an API interface. The OVN scheduling module sends a request to the OVN controller, and the OVN controller returns global memory status information. This global memory status information includes the memory usage of nodes in various types of sandboxes. It should be noted that a unified scheduling strategy could be to prioritize allocating memory to high-priority sandboxes or to perform load balancing across multiple nodes to avoid excessive memory usage on any particular node. In some preferred implementations, the memory scheduling process combined with sandbox business logic includes: The system initiates the sandbox creation process, and the real-time monitoring module synchronously obtains the current node's memory status information and records it to the OVN controller. Select and confirm the use of node1. The OVN controller will make an initial memory allocation plan based on the current memory usage of node1 and the sandbox memory requirements. It should be noted that the specific criteria for selecting and confirming the use of node1 can be based on factors such as the node's remaining memory, CPU utilization, and network bandwidth; for example, the node with the most remaining memory can be selected first, or the node with the closest geographical location can be selected based on the business's location. Nova sends a node occupancy notification to the OVN controller, which then updates the memory resource status of node1 and adjusts the memory scheduling strategy accordingly. Simultaneously, the OVN controller detects changes in memory resources that are about to be occupied, providing data support for subsequent operations. It should be noted that memory scheduling strategies can be adjusted by using rule-based algorithms or machine learning algorithms. For example, rule-based algorithms can formulate rules based on factors such as node memory usage and sandbox priority, and adjust the memory allocation strategy when a certain rule is met. Machine learning algorithms can predict future memory needs by learning from historical data, thereby dynamically adjusting the memory scheduling strategy. The kubelet is taken offline via the OVN controller. During this process, the OVN controller tracks memory release and provides feedback to the OVN controller, which then adjusts memory allocation based on the memory changes. After Nova completes virtual machine creation, it obtains the remaining resource information and updates it to the OVN controller. The intelligent allocation module then uses the remaining resource information, combined with the sandbox and global memory requirements, to re-plan memory allocation. It should be noted that the intelligent allocation module includes an algorithm for replanning memory allocation, which can use either a greedy algorithm or a dynamic programming algorithm. The greedy algorithm selects the optimal allocation scheme at each step to achieve a better overall result; while the dynamic programming algorithm decomposes the problem into subproblems, solves the optimal solutions to the subproblems, and thus obtains the optimal solution to the original problem. The kubelet is brought online through the OVN controller to obtain the latest remaining resource information and update the OVN controller with the latest remaining resource information. The OVN controller further coordinates memory resource allocation based on the latest remaining resource information to ensure the rationality and efficiency of memory use. In some preferred implementations, the memory scheduling process combined with Kubernetes business logic includes: When K8s starts the sandbox creation process, the OVN controller obtains the current node's memory state information and records it to the OVN controller. Select and confirm the use of node1. The OVN controller will make an initial memory allocation plan based on the current memory usage of node1 and the sandbox memory requirements. The kubelet sends a node occupancy notification to the OVN controller, which then updates the memory resource status of node1 and adjusts the memory scheduling strategy accordingly. Simultaneously, the OVN controller monitors changes in memory resource usage to provide data support for subsequent operations. The OVN controller performs the offline nova operation, tracks the memory release status, and feeds back to the intelligent allocation module. The OVN controller further adjusts the memory allocation based on the memory changes. After the kubelet completes the sandbox creation, it obtains the remaining resource information and updates the remaining resource information to the OVN controller. Based on the remaining resource information, combined with the sandbox and global memory requirements, the OVN controller re-plans memory allocation.

[0022] The kubelet is brought online through the OVN controller to obtain the latest remaining resource information and update the OVN controller with the latest remaining resource information; the OVN controller then coordinates memory resource allocation again based on the latest remaining resource information to ensure the rationality and efficiency of memory use; It should be further explained that the specific implementation process of memory scheduling combined with K8s business logic is similar to that of memory scheduling in sandbox business logic, except that kubelet performs the relevant operations on behalf of nova when node occupancy notification and resource creation completion notification are received. In this embodiment of the invention, the above technical solution solves the problem of memory conflict on shared nodes of Kubernetes and OpenStack in the sandbox platform, which is conducive to achieving precise scheduling, improving memory utilization efficiency, dynamically adapting to changes in business load, improving overall system performance, enhancing system stability, and ensuring business continuity.

[0023] Example 2, as Figure 2 As shown, this invention is a sandbox platform technology system for Kubernetes and OpenStack shared nodes based on dynamic memory resource scheduling, comprising: The OVN scheduling module is used to obtain global memory status information from the OVN controller and to coordinate the memory allocation of the OpenStack sandbox and the Kubernetes (K8s) sandbox based on a unified scheduling strategy. The real-time monitoring module is used to synchronously acquire the current node's memory status information when the sandbox creation process starts and record it to the OVN controller; The OVN controller is used to initially plan memory allocation based on node memory usage and sandbox memory requirements; adjust memory scheduling strategies by updating node memory resource status, track memory release, and coordinate memory resource allocation; The intelligent allocation module is used to re-plan memory allocation based on the latest remaining resource information, combined with sandbox and global memory requirements; In some preferred implementations, nova is a computing service component in the OpenStack cloud computing platform, used to send node occupancy notifications to the OVN controller, create virtual machines, and update the remaining resource information to the OVN controller; Kubelet is a node agent in a Kubernetes cluster, used to send node occupancy notifications to the OVN controller, create sandboxes, and update the OVN controller with remaining resource information. Kubernetes (K8s) is an open-source container orchestration platform used to automate the deployment, scaling, and management of containerized applications.

[0024] In the several embodiments provided by this invention, it should be understood that the disclosed system can be implemented in other ways. For example, the embodiments of the invention described above are merely illustrative; for example, the division of modules is only a logical functional division, and there may be other division methods in actual implementation.

[0025] The modules described as separate components may or may not be physically separate. The components shown as modules 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 can be selected to achieve the purpose of this embodiment according to actual needs.

[0026] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0027] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0028] 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 it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for a Kubernetes and OpenStack shared node sandbox platform based on dynamic memory resource scheduling, characterized in that, The method comprises the following steps: Obtaining global memory state information from the OVN controller, and based on a unified scheduling strategy, cooperatively managing memory allocation of OpenStack sandboxes and Kubernetes sandboxes, including a memory scheduling process combined with sandbox business logic and a memory scheduling process combined with K8s business logic; wherein the global memory state information includes node memory usage of each type of sandbox; Synchronously obtaining current node memory state information when the sandbox creation process is started, and recording the same to the OVN controller; According to node memory usage and sandbox memory demand, initially planning memory allocation; by updating node memory resource state, adjusting memory scheduling strategy, tracking memory release, and coordinating memory resource allocation; According to the latest remaining resource information, combined with sandbox and global memory demand, re-planning memory allocation. 2.The Kubernetes and OpenStack shared node sandbox platform technology method based on dynamic memory resource scheduling according to claim 1, wherein, The memory scheduling process combined with sandbox business logic comprises: The system starts the sandbox creation process, and the real-time monitoring module synchronously obtains current node memory state information and records the same to the OVN controller; The node1 node is selected and determined to be used, and the OVN controller initially plans memory allocation according to current node1 node memory usage and sandbox memory demand; The node occupation notification is sent to the OVN controller by using nova, the OVN controller updates node1 node memory resource state, and adjusts the memory scheduling strategy according to the new state; at the same time, the OVN controller monitors the change of memory resource occupation. 3.The Kubernetes and OpenStack shared node sandbox platform technology method based on dynamic memory resource scheduling according to claim 2, wherein, The memory scheduling process combined with sandbox business logic further comprises: The offline kubelet operation is executed by the OVN controller, in the process, the OVN controller tracks the memory release and feeds back to the OVN controller, and the OVN controller further adjusts the memory allocation according to the memory change; After the virtual machine creation is completed by nova, the remaining resource information is obtained, and the remaining resource information is updated to the OVN controller; the intelligent allocation module re-plans memory allocation according to the remaining resource information, combined with sandbox and global memory demand; The kubelet is online by the OVN controller, the latest remaining resource information is obtained, and the latest remaining resource information is updated to the OVN controller; the OVN controller further coordinates memory resource allocation according to the latest remaining resource information. 4.The Kubernetes and OpenStack shared node sandbox platform technology method based on dynamic memory resource scheduling according to claim 1, wherein, The memory scheduling process combined with K8s business logic comprises: The K8s starts the sandbox creation process, and the OVN controller obtains current node memory state information and records the same to the OVN controller; The node1 node is selected and determined to be used, and the OVN controller initially plans memory allocation according to current node1 node memory usage and sandbox memory demand; The node occupation notification is sent to the OVN controller by using kubelet, the OVN controller updates node1 node memory resource state, and the OVN controller adjusts the memory scheduling strategy according to the new state; at the same time, the OVN controller monitors the change of memory resource occupation. 5.The Kubernetes and OpenStack shared node sandbox platform technology method based on dynamic memory resource scheduling according to claim 4, wherein, The memory scheduling process combined with K8s business logic further comprises: The offline nova operation is performed by the OVN controller, the OVN controller tracks the memory release, and feeds back to the intelligent allocation module, and the OVN controller further adjusts the memory allocation according to the memory change; After the kubelet completes the sandbox creation, the remaining resource information is obtained, and the remaining resource information is updated to the OVN controller; the OVN controller re-plans the memory allocation according to the remaining resource information, combined with the sandbox and global memory demand; Through the OVN controller online kubelet, the latest remaining resource information is obtained, and the latest remaining resource information is updated to the OVN controller; the OVN controller adjusts the memory resource allocation again according to the latest remaining resource information.

6. A sandbox platform technology system for Kubernetes and OpenStack shared node based on dynamic memory resource scheduling, applying the sandbox platform technology method for Kubernetes and OpenStack shared node based on dynamic memory resource scheduling according to any one of claims 1-5, characterized in that, Comprise: O The VN scheduling module is used to obtain the global memory state information from the OVN controller, and cooperatively manage the memory allocation of the OpenStack sandbox and the Kubernetes sandbox based on a unified scheduling strategy; The real-time monitoring module is used to synchronously obtain the current node memory state information when the sandbox creation process is started, and record it to the OVN controller; The OVN controller is used to preliminarily plan the memory allocation according to the node memory usage and the sandbox memory demand; By updating the node memory resource state, adjusting the memory scheduling strategy, tracking the memory release, and coordinating the memory resource allocation; The intelligent allocation module is used to re-plan the memory allocation according to the latest remaining resource information, combined with the sandbox and global memory demand. 7.The Kubernetes and OpenStack shared node sandbox platform technology system based on dynamic memory resource scheduling according to claim 6, wherein, Nova is used to send node occupation notification to the OVN controller, create a virtual machine, and update the remaining resource information to the OVN controller. 8.The Kubernetes and OpenStack shared node sandbox platform technology system based on dynamic memory resource scheduling according to claim 7, wherein, Kubelet is used to send node occupation notification to the OVN controller, create a sandbox, and update the remaining resource information to the OVN controller.

Images