Satellite cloud-oriented resource virtualization management system

Abstract

The invention provides a satellite cloud-oriented resource virtualization management system, which comprises a mixed granularity virtualization container, a management and monitoring module and a satellite visualization simulation module, the mixed granularity virtualization container comprises a system-level virtualization container and a container-level virtualization container; the management and monitoring module is used for carrying out full-life-cycle management and resource state monitoring on the system-level virtualization container and the container-level virtualization container; and the satellite visual simulation module is arranged in the container and is used for realizing visual simulation of the satellite according to the relevant parameters of the satellite. The system has container-level and system-level virtualization functions, supports full-life-cycle management and resource state monitoring of containers, realizes standardized packaging, deployment and centralized management of services, provides quick construction of mirror image templates, and carries out fine management and control on various information of mirror images; a visual simulation function is provided for an on-orbit satellite, and the position information of the satellite in a required time period is simulated and observed.

Description

Technical Field

[0001] The present invention relates to the field of satellite technology, and in particular to a resource virtualization management system for satellite clouds. Background Art

[0002] As some companies are accelerating the construction of low-orbit satellite constellations, using the computing power of satellites to achieve on-orbit computing of some high-timeliness data has become a hot topic of current research. Traditional on-board computing environments are mainly customized for fixed tasks and have high reliability requirements, but lack the ability to support general computing, multi-task resource reuse, and cloud-edge collaboration.

[0003] Currently, satellite constellations that are being planned and constructed, such as StarLink, OneWeb, TeleSat, and Sputnix, have shown different characteristics from the already built constellations. These emerging constellations are expected to provide faster Internet communication services to users around the world at a lower cost and with less latency. The construction and development of these emerging satellite constellations will have a profound impact and change on the global communications industry.

[0004] With the rapid development of satellite computing and cloud-native technologies, the interconnection capabilities of networked satellites are becoming increasingly apparent, and the prospect of combining satellite networking with cloud computing technology has attracted much attention. With the continuous advancement of technology and the continuous expansion of application scenarios, we have reason to believe that the deep integration of satellites and cloud computing will bring revolutionary changes to the future communications and computing fields.

[0005] When building satellite-to-ground systems involving satellite interactions, 3D visualization becomes the key to improving system understanding and management efficiency. The Cesium.js framework is a popular open source JavaScript library that focuses on rendering 3D earth and maps in web browsers. This framework provides a rich set of features that allows developers to easily integrate and display various geospatial data, such as 3D models, high-resolution images, terrain elevation data, and vector layers. The power of Cesium.js lies in its ability to handle large-scale data sets and achieve smooth, high-quality rendering on a variety of devices and browsers.

[0006] Based on the above technical background, low-orbit computing, satellite simulation computing and virtualization-related research are developing in the direction of low cost and simplification, and some scholars have designed and implemented related systems. At the same time, we can also see that although the development of these technologies has been relatively mature, they are relatively fragmented in practical applications. Currently, there is no integrated system that fully implements all the above functions. Therefore, the realization of a virtualized management system for satellite-borne edge node resources is not only feasible in terms of technical implementation, but also urgently needed. Summary of the invention

[0007] The technical problem to be solved by the present invention is: based on the integrated low-orbit computing, satellite simulation computing and virtualization related research, the on-board computing resources are integrated to form a unified management view of satellite and ground resources; in view of this, the present invention provides a resource virtualization management system for satellite clouds.

[0008] The technical solution adopted by the present invention is a resource virtualization management system for satellite cloud, comprising:

[0009] Hybrid granularity virtualization containers, including system-level virtualization containers and container-level virtualization containers;

[0010] A management and monitoring module performs full life cycle management and resource status monitoring on the system-level virtualized container and the container-level virtualized container;

[0011] The satellite visualization simulation module is arranged in the container and is used to realize visualization simulation of the satellite according to relevant parameters of the satellite.

[0012] In one embodiment, the satellite visualization simulation is implemented through the sgp4 library and the Cesium.js framework.

[0013] In one embodiment, a satellite constellation configuration to be simulated is obtained, where the satellite constellation configuration is determined by four parameters: the number of constellation orbital planes, the number of satellites per orbit, the orbital inclination, and the orbital altitude;

[0014] Generate the initial constellation configuration according to relevant parameters, and generate the position parameters TLE of the initial satellite;

[0015] Determine the simulation time period and divide the simulation time period into several moments;

[0016] Use the sgp4 library to determine the satellite position information at each moment based on the initial moment TLE and the target moment;

[0017] Assembling the location information into attributes of a satellite entity in a CZML file;

[0018] Send the file to Cesium.js for simulation modeling, and realize the process from TLE to visual simulation.

[0019] In one embodiment, the system-level virtualization container is implemented based on an independent kernel of a virtual machine; and the container-level virtualization container is implemented based on a Namespace and cgroup mechanism of a host operating system kernel.

[0020] In one embodiment, the management and monitoring module is used to implement the following processing for the container: creation, configuration, startup, operation, stop, and destruction; the management and monitoring module is used to collect and monitor the status of the following resources: CPU, memory, and network resource usage of the container.

[0021] In one embodiment, the system is further used to: create the system-level virtualization container and / or the container-level virtualization container based on k3s.

[0022] Another aspect of the present invention provides an electronic device, which includes a resource virtualization management system for a satellite cloud as described in any of the above items.

[0023] Compared with the prior art, the present invention has at least the following advantages:

[0024] The resource virtualization management system for satellite cloud provided by the present invention will have the functions of container-level and system-level virtualization, support container full life cycle management and resource status monitoring, realize standardized encapsulation, deployment and centralized management of services, provide rapid construction of image templates, and perform fine management and control of various information of images. Provide cluster management functions for satellite-borne edge node servers, realize resource scheduling, service deployment, monitoring, security and other management functions for satellite-borne edge node servers, and provide support and monitoring functions for the normal operation of satellite-borne edge node server clusters. Provide visual simulation functions for on-orbit satellites, simulate and observe the location information of satellites in the required time period. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of a container structure according to an embodiment of the present invention;

[0026] Figure 2 A schematic diagram of a container creation process according to an embodiment of the present invention;

[0027] Figure 3 Schematic diagram of resource status management and monitoring logic according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose, the present invention is described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0029] Unless otherwise defined, all terms (including technical terms and scientific terms) used in this article have the same meaning as commonly understood by ordinary technicians in the field to which this application belongs. It should also be understood that terms (such as terms defined in commonly used dictionaries) should be interpreted as having the same meaning as their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined in this article.

[0030] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0031] An embodiment of the present invention provides a resource virtualization management system for satellite clouds, comprising:

[0032] Hybrid granularity virtualization containers, including system-level virtualization containers and container-level virtualization containers;

[0033] A management and monitoring module performs full life cycle management and resource status monitoring on the system-level virtualized container and the container-level virtualized container;

[0034] The satellite visualization simulation module is arranged in the container and is used to realize visualization simulation of the satellite according to relevant parameters of the satellite.

[0035] In this embodiment, satellite visualization simulation is implemented through the sgp4 library and the Cesium.js framework.

[0036] In this embodiment, a satellite constellation configuration to be simulated is obtained, and the satellite constellation configuration is determined by four parameters: the number of constellation orbital planes, the number of satellites per orbit, the orbital inclination, and the orbital altitude;

[0037] Generate the initial constellation configuration according to relevant parameters, and generate the position parameters TLE of the initial satellite;

[0038] Determine the simulation time period and divide the simulation time period into several moments;

[0039] Use the sgp4 library to determine the satellite position information at each moment based on the initial moment TLE and the target moment;

[0040] Assembling the location information into attributes of a satellite entity in a CZML file;

[0041] Send the file to Cesium.js for simulation modeling, and realize the process from TLE to visual simulation.

[0042] In this embodiment, the system-level virtualization container is implemented based on the independent kernel of the virtual machine; the container-level virtualization container is implemented based on the Namespace and cgroup mechanism of the host operating system kernel.

[0043] In this embodiment, the management and monitoring module is used to implement the following processing for the container: creation, configuration, startup, operation, stop, and destruction; the management and monitoring module is used to collect and monitor the status of the following resources: CPU, memory, and network resource usage of the container.

[0044] In this embodiment, the system is further used to: create the system-level virtualization container and / or the container-level virtualization container based on k3s.

[0045] The system provided in this embodiment will be described in detail below.

[0046] 1. Hybrid-granularity virtualization containers

[0047] refer to Figure 1 ,Hybrid-granularity virtualization containers include system-level virtualization containers and container-level virtualization containers, which use system-level virtualization technology and container-level virtualization technology respectively. The main research result of this part is to integrate the two containers into k3s.

[0048] The basic principle of system-level virtualization is to start a virtual machine for each container (if it is used as a runtime under k3s, one pod corresponds to one virtual machine instead of a container). Each virtual machine has an independent kernel, so that the delivered container has virtual machine-level isolation and security. Figure 1 As shown in the figure, the virtualization principle of container-level virtualization is to use the cgroup and namespace mechanism of the host operating system kernel. The container created in this way has an execution environment that is isolated from other process resources in the system. It is shared with the host kernel. The application in the container can circumvent the resource restrictions of the cgroup on the container or even obtain the host permissions through system calls, or cause the container to access memory out of bounds through vulnerabilities in the operating system kernel.

[0049] System-level virtualization combines containers with virtual machine (VM) technology, builds an operating system kernel for each container, and allows each container to run in its own lightweight virtual machine, thus achieving hardware-level isolation. This isolation can help prevent attacks and escapes between containers and provide higher security protection. However, the cost is poor performance.

[0050] 2. Container Lifecycle Management and Resource Status Monitoring

[0051] The full life cycle management of containers is done through the container orchestration tool k3s to manage two types of virtualized containers. The following is a detailed introduction to the workflow after the integration of virtualized containers and k3s. When users need to create a system-level virtualized container in k3s, such as Figure 2 As shown, the workflow is as follows.

[0052] 1) The kubelet component of k3s uses gRPC that meets the CRI standard to communicate with the high-level container runtime Docker or containerd to request the creation of a system-level virtualized container;

[0053] 2) After receiving the creation request, the high-level container runtime will use ttRPC that meets the shim v2 standard to communicate with the sock socket of containerd-shim-kata-v2 on the host machine to request the creation of a system-level virtualized container;

[0054] 3) containerd-shim-kata-v2 will use ttRPC to send a request to the kata-agent running in the virtual machine. After receiving the request, if kata-agent needs to create a Pod, it will create a virtual machine for this request, and then start several containers in this virtual machine. At this time, the virtual machine corresponds to the Pod in k3s, and all containers in the same virtual machine share a virtual kernel.

[0055] When a user needs to create a container-level virtualized container, the workflow is similar to the system-level virtualized container creation process. The difference is that the underlying container runtime uses the cgroup and namespace mechanisms of the operating system to create the container. At this time, the container and the host share an operating system kernel.

[0056] Through the above method, k3s can realize the process of creating two types of containers, and other operations of container full life cycle management can also be completed through a similar process.

[0057] The system's resource status monitoring is implemented based on Prometheus, such as Figure 3 This is the architecture diagram of Prometheus. It includes the following components:

[0058] 1. Prometheus Server: It is the core component of the system, responsible for collecting and storing time series data, providing HTTP API for querying and retrieving metrics data, and also responsible for data storage;

[0059] 2.Client Library: It is an official library embedded in the user application and is used to collect corresponding metrics data through PrometheusServer;

[0060] 3. Exporters: Programs specifically used to expose metrics data for specific services. Since Prometheus does not natively support metrics collection for some services or applications, exporters can be used to bridge these services and convert metrics data into a format that Prometheus can understand.

[0061] 4.Alertmanager: responsible for processing the warning information sent by Prometheus Server, it can deduplicate and group the warnings, and send the warnings to different recipients (such as email, Slack, webhook, etc.) according to the configured routing rules;

[0062] 5. Grafana: It is an open source monitoring dashboard and visualization tool monitoring dashboard that can query metrics data through the API provided by Prometheus and display it to users in a graphical way;

[0063] 6. pushgateway: It allows short-running tasks or services to push metrics data to Pushgateway instead of being pulled by Prometheus Server like regular services.

[0064] Its main workflow is as follows:

[0065] 1) Data collection: Prometheus server regularly pulls monitoring data from the host to be monitored;

[0066] 2) Data storage: Prometheus Server stores the collected monitoring indicator data in time series in its built-in database. This data can be used for subsequent analysis, query and alarm;

[0067] 3) Alarm and notification: Prometheus Server can trigger alarms based on thresholds or conditions of monitoring data by configuring alarm rules. When an alarm is triggered, Prometheus Server will send the alarm information to the Alertmanager component. The Alertmanager is responsible for processing the alarm information. It can deduplicate and group the alarm information according to the configuration and send the alarm information to the specified recipient, such as email, Slack notification, webhook, etc.

[0068] 4) Data query and visualization: Prometheus Server provides a Web UI interface, where users can query and retrieve monitoring data through the PromQL (Prometheus Query Language) query language.

[0069] Through the above method, the system's resource monitoring data can be exposed through the Prometheus webui, and the relevant data can be obtained using the PromQL query language through the corresponding interface.

[0070] 3. Satellite Visualization Simulation

[0071] Satellite visualization simulation is implemented through the sgp4 library and Cesium.js framework. The sgp4 library can accept orbit information encoded in TLE format. After the initial time information is given, the geographical location of the satellite at a certain time after any event can be calculated. Then this information can be exported as CZML format data and handed over to Cesium.js for visualization.

[0072] CZML (Cesium ZML) is a JSON-based data format used by Cesium.js to describe entities and their properties in 3D scenes. In CZML, a satellite or other object is called an "entity", and the properties of an entity are represented by a dictionary (or a set of key-value pairs). The properties of a CZML entity describing a satellite orbit can be seen in Table 1.

[0073] Table 1

[0074]

[0075] When it is necessary to perform a visual simulation of a satellite constellation, the satellite constellation configuration to be simulated must first be given. This can usually be determined by four parameters: the number of constellation orbital planes, the number of satellites per orbit, the orbital inclination, and the orbital altitude. After determining the satellite constellation configuration, the initial constellation configuration can be generated based on the relevant parameters, and the position parameters of the initial satellite, that is, TLE, can be generated. After obtaining the initial TLE of each satellite, the simulation time period must be obtained, and then the simulation time period is divided into several moments. The sgp4 library is used to calculate the satellite position information at each moment based on the initial moment TLE and the target moment, and then this information is assembled into the attributes of the satellite entity in the CZML file, and the file is sent to Cesium.js for simulation modeling, thus realizing the process from TLE to visual simulation.

[0076] Compared with the prior art, this embodiment has at least the following effects:

[0077] 1) The present invention provides a hybrid granularity virtualization container: The hybrid granularity virtualization container is mainly aimed at the heterogeneous problem of satellite-borne edge nodes. The edge nodes are composed of a large number of heterogeneous nodes. In order to integrate the computing resources on the edge nodes and realize the isolation between different services, it is necessary to realize virtualization on the heterogeneous nodes. The system can build system-level virtualization containers and container-level virtualization containers, wherein the system-level virtualization is implemented based on the independent kernel of the virtual machine, with better security and isolation; the container-level virtualization is implemented based on the Namespace and cgroup mechanism of the host operating system kernel, with lower resource consumption and shorter startup time.

[0078] 2) The present invention realizes container full life cycle management and resource monitoring: Based on the implementation of container-level and system-level virtualization, in order to better monitor the performance indicators of the container and improve the performance and availability of the system, further research is needed to realize container full life cycle management and resource status monitoring. Container full life cycle management includes all stages from container creation to destruction, including operations such as container creation, configuration, startup, operation, stop, and destruction. Resource status monitoring includes collecting and analyzing the container's CPU, memory, network and other resource usage.

[0079] 3) The present invention can be used for satellite visualization simulation: the system can accurately calculate the satellite's position information within a specific time period by integrating high-precision orbital data and initial position information. This information is then presented on a dynamic three-dimensional earth model, allowing users to intuitively observe the precise position of the satellite at each moment.

[0080] The second embodiment of the present invention is an electronic device, which can be understood as a physical device, including a resource virtualization management system for a satellite cloud as provided in the first embodiment.

[0081] Through the description of the specific implementation methods, a deeper and more specific understanding of the technical means and effects adopted by the present invention to achieve the predetermined purpose should be obtained. However, the accompanying drawings are only for reference and illustration purposes and are not intended to limit the present invention.

Claims

1. A resource virtualization management system for satellite cloud, characterized in that: include: Hybrid granularity virtualization containers, including system-level virtualization containers and container-level virtualization containers; A management and monitoring module performs full life cycle management and resource status monitoring on the system-level virtualized container and the container-level virtualized container; The satellite visualization simulation module is arranged in the container and is used to realize visualization simulation of the satellite according to relevant parameters of the satellite.

2. The resource virtualization management system for satellite cloud according to claim 1, characterized in that: The satellite visualization simulation is implemented through the sgp4 library and the Cesium.js framework.

3. The resource virtualization management system for satellite cloud according to claim 2, characterized in that: The visual simulation of the satellite includes: Obtaining a satellite constellation configuration to be simulated, wherein the satellite constellation configuration is determined by four parameters: the number of constellation orbital planes, the number of satellites per orbit, the orbital inclination, and the orbital altitude; Generate the initial constellation configuration according to relevant parameters, and generate the position parameters TLE of the initial satellite; Determine the simulation time period and divide the simulation time period into several moments; Use the sgp4 library to determine the satellite position information at each moment based on the initial moment TLE and the target moment; Assembling the location information into attributes of a satellite entity in a CZML file; Send the file to Cesium.js for simulation modeling, and realize the process from TLE to visual simulation.

4. The resource virtualization management system for satellite cloud according to claim 1, characterized in that: The system-level virtualization container is implemented based on the independent kernel of the virtual machine; the container-level virtualization container is implemented based on the Namespace and cgroup mechanism of the host operating system kernel.

5. The resource virtualization management system for satellite cloud according to claim 1, characterized in that: The management and monitoring module is used to implement the following processing for the container: creation, configuration, startup, operation, stop, and destruction; the management and monitoring module is used to collect and monitor the status of the following resources: CPU, memory, and network resource usage of the container.

6. The resource virtualization management system for satellite cloud according to claim 1, characterized in that: The system is further used to: create the system-level virtualization container and / or the container-level virtualization container based on k3s.

7. An electronic device, characterized in that: The electronic device includes a satellite cloud-oriented resource virtualization management system as described in any one of claims 1 to 6.

Images