Network function virtualization (NFV) system and resource management method

Abstract

The embodiment of the invention discloses an NFV (Network Function Virtualization) system and a resource management method, the system comprises a management node and an NFVI (Network Function Virtualization Infrastructure) node, and at least two VMs (Virtual Machine) run in the NFVI node; the NFVI node is used for receiving a first request from the management node, the first request comprises information of a target VM, and the target VM is included in at least two VMs; the NFVI node is used for mapping read-only data of a shared memory space into a file path of a target VM according to information of the target VM, the read-only data is operating system data and / or container mirror image data shared by at least two VMs, the operating system data is used for initializing the target VM, and the container mirror image data is used for enabling the target VM to run a service corresponding to the container mirror image data; the target VM is used for initializing and / or running a service corresponding to the container mirror image data according to the operating system data. According to the technical scheme, the utilization rate of memory resources can be increased.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of communication technologies, and in particular, to a network function virtualization (NFV) system and a resource management method. Background Art

[0002] Network Function Virtualization (NFV) mainly formulates industry-wide general standards for network cloudification. Based on modern IT virtualization technologies, it provides a new network product environment, reduces costs, improves efficiency, and increases agility. By using general-purpose hardware such as x86 and virtualization technologies to carry out software processing of many functions, the expensive equipment costs of the network are reduced.

[0003] In the NFV architecture, a commonly used technology is the storage bypass technology, which places the data in the storage device in the memory of the physical machine and allows the virtual machine (VM) to directly access the data without additional I / O processing. This improves data transmission efficiency and reduces the risk of service interruption due to storage failures.

[0004] Due to the isolation characteristics of virtualization technologies, when multiple VMs are running on the same physical machine, the hardware resources used by the VMs are isolated. Therefore, the guest operating system (Guest OS) and the container images used by each VM are stored in the memory of the physical machine. At this time, the read-only data in the Guest OS and the container images is repeatedly loaded in the memory of the physical machine, thus occupying more memory space. Summary of the Invention

[0005] The embodiments of the present application provide a network function virtualization (NFV) system and a resource management method for improving the utilization rate of memory resources. The embodiments of the present application also provide corresponding resource management devices.

[0006] In a first aspect, an embodiment of the present application provides a network function virtualization (NFV) system, which includes a management node and a network functions virtualization infrastructure (NFVI) node. At least two virtual machines (VMs) are running in the NFVI node. The NFVI node is configured to receive a first request from the management node, where the first request carries information about a target VM, and the target VM is included in the at least two VMs running in the NFVI. The NFVI node is further configured to map the read-only data in the shared memory space to the file path of the target VM according to the information of the target VM. The read-only data is operating system data and / or container image data shared by the at least two VMs. The operating system data is used to initialize the target VM, and the container image data is used for the target VM to run the service corresponding to the container image data. The target VM in the NFV system is configured to be initialized according to the operating system data and / or run the service corresponding to the container image data.

[0007] In the present application, the NFVI node is used to provide the virtualization resources required to support the execution of the NFV system, which is a software layer that virtualizes and abstracts the underlying hardware. The NFVI node includes a hardware resource layer composed of computing hardware, network hardware, and storage hardware, a virtualization layer, and a virtual resource layer composed of virtual computing, virtual storage, and virtual network. The management node is each network element node in the network function virtualization management and orchestration (NFV-MANO). A VM is the carrier of a virtualized network function (VNF) on the NFVI node and is responsible for running the software environment of the VNF. Among them, the operating system data of the VM is used for system initialization, and the container image data of the VM is used for the software service to implement the VNF function.

[0008] By adopting the above method, the shared memory space stores the read-only data shared by at least two VMs. Since each VM runs using the memory resources of the HOST, extracting the read-only data and storing it in the shared memory space can save the memory space required for each VM, reduce the repeated download and repeated occupation of memory space by data in different VMs, and improve the utilization rate of memory resources. Further, the saved memory space can also be used to expand more new VMs. In addition, mapping the read-only data to the file path of the target VM can improve the startup and loading speed.

[0009] In some alternative embodiments, when the read-only data is operating system data shared by at least two VMs, the management node is a virtual network function manager (VNFM) node at this time. The target VM is specifically configured to: merge the shared operating system data with the first data to generate a first directory, where the first data is the read-write data of the operating system of the target VM and is stored in the memory space of the target VM; the target VM initializes based on the first directory.

[0010] In this application, a possible implementation is that the directories corresponding to the read-only data in the shared memory disk and the directories corresponding to the read-write data in its own memory space can be merged through overlayfs. By adopting the above method, the collection and collation of data are completed at one time by using the directory merging operation, reducing subsequent repeated operations on the data and improving the efficiency.

[0011] In some alternative embodiments, when the read-only data is container image data shared by at least two VMs, the management node is a container cluster management (CCM) node at this time. The target VM is specifically configured to: run the service corresponding to the container image data based on the container metadata and the shared container image data, and the container metadata is stored in the memory space of the target VM.

[0012] In this application, the CCM node can manage the CIS module on the VM. First, the container data of the VM is divided into container image layers (snapshots) and container metadata. The snapshots are stored in the shared memory space, and then a soft link is established for access when the VM needs to use the container image, and the usage count of the corresponding container image is recorded.

[0013] In some alternative embodiments, the NFVI node is further configured to: receive a second request from the management node, where the second request includes an identifier of the shared memory space; establish an association between the shared memory space and each of at least two VMs.

[0014] In this application, the management node further includes a Network Function Virtualization Orchestrator (NFVO) node, a Virtual Network Function Manager (VNFM), and a Virtualized Infrastructure Manager (VIM) node. In a possible implementation manner, the specific process for the management node to send the second request includes: the NFVO node sends the second request to the VNFM node, the VNFM node forwards the request to the VIM node, and the VIM node forwards the request to the NFVI node. By using the above method, by mounting a shared memory space to the VM and directly accessing the data in the memory space, network transmission and disk I / O operations are reduced, thereby improving the overall performance of the system.

[0015] In some alternative embodiments, the NFVI node is further configured to: receive a third request from the management node, where the third request is used to instruct the NFVI node to create a shared memory space, and the third request includes the memory size to be occupied; determine a memory space that meets the occupied memory size from the virtualized memory space as the shared memory space.

[0016] In some alternative embodiments, the management node includes a Network Function Virtualization Orchestrator NFVO node, a Virtual Network Function Manager VNFM node, and a Virtualized Infrastructure Manager VIM node; the VIM node is configured to send a third request to the NFVI node according to a fourth request; the NFVO node is configured to send the fourth request to the VIM node or the VNFM node; the VNFM node is configured to forward the fourth request to the VIM node.

[0017] In this application, the process for the management node to send a creation request may be based on the NFV lifecycle management process, sent by the NFVO node to the VNFM node and then forwarded to the VIM node; or it may directly manage the virtualized resources, sent directly by the NFVO node to the VIM node. After receiving the creation request, the VIM node forwards the creation request to the NFVI node, instructing the NFVI node to create the shared memory space.

[0018] Second aspect, an embodiment of the present application provides a resource management method, which is applied to a network function virtualization (NFV) system. The NFV system includes a management node and a network function virtualization infrastructure (NFVI) node. At least two virtual machines (VMs) are running in the NFVI node. The method includes: The NFVI node receives a first request from the management node. The first request includes information about a target VM, and the target VM is included in the at least two VMs. The NFVI node maps the read-only data in the shared memory space to the file path of the target VM according to the information of the target VM. The read-only data is operating system data and / or container image data shared by the at least two VMs. The operating system data is used to start the target VM, and the container image data is used for the target VM to run the service corresponding to the container image data.

[0019] By adopting the above method, the read-only data shared by at least two VMs is stored in the shared memory space. Since each VM runs using the memory resources of the HOST, extracting the read-only data and storing it in the shared memory space can save the memory space required by each VM, reduce the repeated download and repeated occupation of memory space by the data in different VMs, and improve the utilization rate of memory resources. Mapping the read-only data to the file path of the target VM can improve the startup and loading speed.

[0020] In some optional embodiments, the management node is a virtual network function manager (VNFM) node or a container cluster manager (CCM) node.

[0021] In some optional embodiments, before the NFVI node receives the first request from the management node, the method further includes: The NFVI node receives a second request from the management node. The second request includes an identifier of the shared memory space. The NFVI node establishes an association between the shared memory space and each of the at least two VMs according to the identifier of the shared memory space.

[0022] In some optional embodiments, before the NFVI node receives the second request from the management node, the method further includes: The NFVI node receives a third request from the management node. The third request is used to instruct the NFVI node to create a shared memory space, and the third request includes the occupied memory size. The NFVI node determines a memory space that meets the occupied memory size from the virtualized memory space as the shared memory space according to the third request.

[0023] In some optional embodiments, the management node includes a network function virtualization orchestrator (NFVO) node, a virtual network function manager (VNFM) node, and a virtualization infrastructure manager (VIM) node. The third request is generated by the VIM node according to a fourth request. The fourth request is sent by the NFVO node to the VIM node or the VNFM node forwards the fourth request of the NFVO to the VIM node.

[0024] In a third aspect, an embodiment of the present application provides a resource management method, which is applied to a network function virtualization (NFV) system. The NFV system includes a network function virtualization infrastructure (NFVI) node, and at least two virtual machines (VMs) are running in the NFVI node. The at least two VMs are each associated with a shared memory space. The method includes: a target VM obtains read-only data in the shared memory space, where the read-only data is operating system data and / or container image data shared by the at least two VMs, and the target VM is one of the at least two VMs; the target VM initializes and / or runs a service corresponding to the container image data according to the operating system data.

[0025] In the present application, through the shared memory space, multiple VMs can share the operating system data and container image data, avoiding the repeated occupation and waste of resources and improving the utilization rate of memory resources. In addition, the introduction of the shared memory space enables the NFV system to easily expand new VMs, enhancing the scalability and reliability of the system.

[0026] In some optional embodiments, when the read-only data is operating system data shared by the at least two VMs, the target VM initializes according to the operating system data, including: the target VM merges the shared operating system data with first data to obtain a first directory, where the first data is read-write data of the operating system of the target VM and is stored in the memory space of the target VM; the target VM initializes based on the first directory.

[0027] In some optional embodiments, when the read-only data is container image data shared by the at least two VMs, the target VM runs a service corresponding to the container image data, including: the target VM runs a service corresponding to the container image data based on container metadata and the shared container image data, and the container metadata is stored in the memory space of the target VM.

[0028] In a fourth aspect, an embodiment of the present application provides a resource management device, which is characterized in that it is applied to a network function virtualization (NFV) system. The NFV system includes a management node and a network function virtualization infrastructure (NFVI) node, and at least two virtual machines (VMs) are running in the NFVI node; the resource management device is disposed in the NFVI node, and the resource management device includes: a receiving module, configured to receive a first request from the management node, where the first request includes information of a target VM, and the target VM is included in the at least two VMs; a processing module, configured to map the read-only data in the shared memory space to a file path of the target VM according to the information of the target VM, where the read-only data is operating system data and / or container image data shared by the at least two VMs, the operating system data is used to start the target VM, and the container image data is used for the target VM to run a service corresponding to the container image data.

[0029] In some alternative embodiments, the management node is a Virtual Network Function Manager (VNFM) node or a Container Cluster Manager (CCM) node.

[0030] In some alternative embodiments, before the NFVI node receives a first request from the management node, the receiving module is further configured to receive a second request from the management node, where the second request includes an identifier of a shared memory space; and the processing module is further configured to establish an association between the shared memory space and each of at least two VMs according to the identifier of the shared memory space.

[0031] In some alternative embodiments, before the NFVI node receives a second request from the management node, the receiving module is further configured to receive a third request from the management node, where the third request is used to instruct the NFVI node to create a shared memory space, and the third request includes the occupied memory size; and the processing module is further configured to determine, according to the third request, a memory space that meets the occupied memory size from the virtualized memory space as the shared memory space.

[0032] In some alternative embodiments, the management node includes a Network Function Virtualization Orchestrator (NFVO) node, a Virtual Network Function Manager (VNFM) node, and a Virtualized Infrastructure Manager (VIM) node; the third request is generated by the VIM node according to a fourth request, and the fourth request is sent by the NFVO node to the VIM node or the VNFM node forwards the fourth request of the NFVO to the VIM node.

[0033] In a fifth aspect, an embodiment of the present application provides a resource management apparatus, which is characterized in that it is applied to a Network Function Virtualization (NFV) system. The NFV system includes a Network Function Virtualization Infrastructure (NFVI) node, and at least two Virtual Machines (VMs) are running in the NFVI node. Each of the at least two VMs is associated with a shared memory space; the resource management apparatus is disposed in a target VM, and the target VM is one of the at least two VMs. The resource management apparatus includes: a receiving module, configured to obtain read-only data in the shared memory space, where the read-only data is operating system data and / or container image data shared by the at least two VMs; and a processing module, configured to perform initialization and / or run services corresponding to the container image data according to the operating system data.

[0034] In some alternative embodiments, when the read-only data is operating system data shared by the at least two VMs, the processing module is specifically configured to: merge the shared operating system data with first data to obtain a first directory, where the first data is read-write data of the operating system of the target VM, and the first data is stored in the memory space of the target VM; and perform initialization based on the first directory.

[0035] In some alternative embodiments, when the read-only data is container image data shared by at least two VMs, the processing module is specifically configured to: run the service corresponding to the container image data based on the container metadata and the shared container image data, where the container metadata is stored in the memory space of the target VM.

[0036] In a sixth aspect, an embodiment of the present application provides a resource management device, which includes: a processor, a memory, and a transceiver. A computer program or computer instructions are stored in the memory, and the processor is configured to call and run the computer program or computer instructions stored in the memory, so that the processor implements the processing operations in the second aspect and any implementation manner in the second aspect. The transceiver is configured to transmit and receive signals, such as: implementing the receiving and transmitting operations in the second aspect and any implementation manner in the second aspect.

[0037] In a seventh aspect, an embodiment of the present application provides a communication device, which includes: a processor, a memory, and a transceiver. A computer program or computer instructions are stored in the memory, and the processor is configured to call and run the computer program or computer instructions stored in the memory, so that the processor implements the processing operations in the third aspect and any implementation manner in the third aspect. The transceiver is configured to transmit and receive signals, such as: implementing the receiving and transmitting operations in the third aspect and any implementation manner in the third aspect.

[0038] In an eighth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program runs on a computer, it causes the computer to execute the second aspect and any optional method in the second aspect.

[0039] In a ninth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program runs on a computer, it causes the computer to execute the third aspect and any optional method in the third aspect.

[0040] In a tenth aspect, an embodiment of the present application provides a computer program. When the computer program runs on a computer, it causes the computer to execute the second aspect and any optional method in the second aspect.

[0041] In an eleventh aspect, an embodiment of the present application provides a computer program. When the computer program runs on a computer, it causes the computer to execute the third aspect and any optional method in the third aspect.

[0042] As described above, the technical effects of the fourth, sixth, eighth, and tenth aspects of this application can be understood in conjunction with the technical effects of the second aspect and any implementation manner of the second aspect. The technical effects of the fifth, seventh, ninth, and eleventh aspects of this application can be understood in conjunction with the technical effects of the third aspect and any implementation manner of the third aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] To more clearly illustrate the technical solutions in the embodiments of this application, the following briefly introduces the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of this application. For those skilled in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0044] Figure 1 Schematic diagram of an NFV system architecture provided by an embodiment of this application;

[0045] Figure 2 Schematic diagram of the data path of the storage bypass technology;

[0046] Figure 3 Schematic flowchart of a resource management method provided by an embodiment of this application;

[0047] Figure 4 Another schematic flowchart of a resource management method provided by an embodiment of this application;

[0048] Figure 5 Schematic diagram of sharing a Guest OS provided by an embodiment of this application;

[0049] Figure 6 Another schematic flowchart of a resource management method provided by an embodiment of this application;

[0050] Figure 7 Schematic diagram of sharing a container image provided by an embodiment of this application;

[0051] Figure 8 Schematic diagram of a structure of a resource management device provided by an embodiment of this application;

[0052] Figure 9 Another schematic diagram of a structure of a resource management device provided by an embodiment of this application;

[0053] Figure 10 Another schematic diagram of a structure of a resource management device provided by an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0055] Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification, claims and the above-mentioned drawings of the present application are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described here, for example, can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0056] To facilitate the understanding of the present invention, the network function virtualization (NFV) system architecture in the related art will be briefly introduced below.

[0057] Virtualization refers to converting a physical computer (HOST) into multiple logical computers through virtualization software. This technology enables a computer to run multiple operating systems simultaneously, and each operating system runs in its own independent space without interference. This not only improves the utilization rate of computer hardware resources but also greatly enhances the working efficiency of the computer. Virtualization uses software methods to redefine and partition IT resources. This virtualization technology can achieve the dynamic allocation and flexible scheduling of IT resources, enabling IT resources to be shared among different fields, thereby more effectively meeting various complex and changing application requirements. Generally speaking, virtualization technology can improve the utilization rate of IT resources and make it better serve various industries and fields.

[0058] Virtualization is the key technology to achieve NFV. NFV enables telecommunications network operators to integrate many types of network devices into a large number of servers, switches, and storage devices that meet industry standards by using general-purpose hardware such as X86 and by researching and developing standard IT virtualization technologies, and execute network functions in the software running on a series of industry-standard server hardware. Here, the software can be installed and uninstalled on the hardware at different positions in the network as needed, without the need to install new hardware devices.

[0059] The operator established the Network Functions Virtualization Industry Specification Group (NFV ISG) in the European Telecommunications Standards Institute (ETSI). Its research goal is mainly to widely adopt standardized IT virtualization technologies, use industry-standard high-capacity servers, storage, and switches to carry various network software functions, achieve flexible software loading, and achieve flexible deployment and configuration at various locations such as data centers, network nodes, and user terminals, so as to accelerate the speed of network deployment and adjustment, reduce the complexity of service deployment, improve the standardization, generalization, and adaptability of network devices, and ultimately reduce the fixed asset investment and operating costs of the network.

[0060] As Figure 1 shown, it shows a schematic diagram of the architecture of the NFV system in the related art. The NFV system includes a Network Functions Virtualization Infrastructure (NFVI), multiple Virtualized Network Functions (VNFs) (such as VNF1, VNF2, and VNF3), one or more Element Managers (EMs) (such as EM1, EM2, and EM3), one or more Operations Support Systems and Business Support Systems (OSS / BSS), and a Network Functions Virtualization Management and Orchestration (NFV-MANO).

[0061] The NFVI is used to provide the virtualized resources required to support the execution of the NFV system, including Commercial Off-The-Shelf (COTS) hardware, necessary accelerator components, and a software layer that virtualizes and abstracts the underlying hardware. For example, the NFVI can include a hardware resource layer composed of computing hardware, network hardware, and storage hardware, a virtualization layer, and a virtual resource layer composed of virtual computing, virtual storage, and virtual networks.

[0062] Among them, the virtualization layer is a software layer composed of software required to host virtual instances, which is installed on top of the hardware layer of the COTS host. The virtualization layer includes but is not limited to: a bootloader, a cloud operating system (CloudOS) or a host operating system (hostOS), a hypervisor, a virtual machine manager (VMM), and other software required for security, transmission, and business to support the operation of virtual instances. This application does not limit them.

[0063] The VNF implements the functions of traditional non-cloudized telecom network elements. The resources required by the VNF need to be decomposed into virtual computing / storage / switching resources and borne by the NFVI. A VNF can be deployed on one or more virtual machines (VMs). Conceptually, the VNF is a virtualized implementation of a network function (NF). There may also be an attached element management system (EMS) to understand and manage individual VNFs and their characteristics. The VNF is equivalent to the entity of a network node and is expected to be delivered as pure software that is independent of hardware. Among them, the Container Infrastructure Service (CIS) is deployed on the VM to execute services.

[0064] A VM is a complete operating system environment that includes a kernel of an operating system, applications, and all necessary libraries. It runs in an isolated environment provided by a software called a virtual machine monitor or hypervisor. Each virtual machine has its own operating system instance and can run independently. In the embodiments of this application, the operating system running on the VM is a guest operating system (Guest OS).

[0065] The CIS is the data plane of the container management platform, which realizes the creation and deletion of containers and container networks by using container technologies (such as Docker or Kubernetes), supports virtual machines or bare metal, and is used to run services.

[0066] A Pod is the smallest control unit of Kubernetes, and containers all run in a Pod. There can be one or more containers in a Pod.

[0067] The EMS, including one or more EMs, is mainly used to perform traditional fault, configuration, user, performance, and security management functions for the VNF.

[0068] OSS / BSS mainly targets at telecom service operators and provides comprehensive network management and business operation functions, including network management (such as fault monitoring, network information collection, etc.), charging management, and customer service management, etc.

[0069] NFV-MANO is a set of automated deployment specifications defined in the NFV-MANO domain according to the requirements of the ETSI NFV ISG specification. It can implement a set of virtualized application deployment processes and architectures, including network function virtualization orchestrator (NFVO), one or more virtual network function managers (VNFM), virtualized infrastructure manager (VIM), container cluster management (CCM), and container infrastructure service management (CISM).

[0070] NFVO is responsible for the orchestration and maintenance of network services, physical / virtual resources, and policies across the entire network, as well as other maintenance management functions related to virtualized systems. It realizes the management of the network service life cycle, and cooperates with VNFM to realize the life cycle management of VNF and the global view function of resources; it can also communicate with VIM to realize resource allocation, and realize the reservation and exchange of configuration information and status information of virtualized hardware resources. NFVO can also interface with OSS / BSS to obtain the service description of network services.

[0071] VNFM is mainly responsible for deploying or managing the corresponding VNF. It should be understood that each VNF is assumed to have an associated VNFM, and a VNFM can be assigned to manage a single VNF instance or manage multiple VNF instances of the same or different types. VNFM is mainly responsible for: VNF instantiation, VNF configuration of NFV resources, VNF instance update, VNF instance scaling, collection of NFVI performance metrics and events related to VNF instances and correlation of events related to VNF instances, assisted or automatic recovery of VNF instances, termination of VNF instances, integrity management of VNF instances throughout their life cycle, acting as a global coordinator and adapter for configuration and event reporting between NFVI and EMS, etc. For example, VNFM can add VNF, delete VNF, search for VNF, or manage VNF according to the requirements of network management, such as monitoring and adjusting the status of VNF.

[0072] The VIM is mainly responsible for controlling the NFVI to provide corresponding virtual resources for the VNF. It should be understood that within the infrastructure sub-domain of an operator, a VIM can specifically handle a certain type of NFV resource or manage multiple types of NFV resources. The VIM is mainly responsible for orchestrating the allocation / upgrade / de-allocation / recycling of NFV resources, managing the association between virtualized resources and computing, storage, and network resources; managing the catalogs of hardware resources (computing, storage, network) and software resources (such as hypervisors); collecting and forwarding performance metrics and events of virtualized resources, etc. For example, the VIM can, according to the scheduling of the NFVO, control the NFVI to provide corresponding virtual resources for the deployment or management of the VNF. The VIM can be a cloud platform, such as an open-source cloud platform like OpenStack, or a commercial cloud platform like VMware.

[0073] The CISM function includes resource discovery, management and scheduling of container objects, etc. It is mainly responsible for managing the container objects invoked by the containerized VNF, including the creation, update, and deletion of container objects, and scheduling the container objects to the corresponding node resources in the container cluster node resource pool managed by the CISM. The CISM is also responsible for managing the process container infrastructure service CIS instances of the container user plane.

[0074] The CCM is responsible for managing the container cluster, including the creation of the node resource pool used by the container cluster and the scaling in and out of nodes. For example, allocating new VMs or bare-metal machines (BMs) in the specified container cluster node resource pool.

[0075] It should be understood that both the CISM and the CCM are independent logical functions. During the functional deployment process, the CCM and the CISM can be deployed in different physical entities. For example, the CISM is deployed in the VNFM and the CCM is deployed in the VIM; they can also be deployed in the same physical entity. For example, both are deployed in the VNFM.

[0076] Based on the aforementioned NFV system, processes such as the deployment, upgrade, and expansion of virtualized network functions (VNF) can be carried out. In existing technical solutions, the storage bypass technology is usually used to achieve direct reading and writing of virtual machine data to improve the I / O performance of the virtual machine. The storage bypass technology can bypass the I / O processing of the virtualization layer, enabling the virtual machine to directly access the storage device, reducing the intermediate links in data transmission, and thus improving the data transmission efficiency.

[0077] Please refer to Figure 2 , Figure 2Schematic diagram of the storage bypass data path. For the storage bypass, to ensure that a remote storage failure does not affect the basic service operation, the Guest OS and the Thinpool directory are placed in the VM memory disk. Before the operating system starts, the Guest OS files are loaded from the remote storage to the memory disk, and the operating system boots from the memory disk. The Thinpool directory is used to store container images and also uses the memory disk to download the container images to the memory disk.

[0078] The storage bypass technology runs the Guest OS and container images in memory. The data stored in memory is less vulnerable to external attacks or accidental data loss compared to traditional disk storage. In case of a storage failure, the network element continues to run in memory and provides a minimum maintenance capability channel, and the storage failure does not affect the service.

[0079] The applicant's research found that when multiple VMs are running on the same physical machine, each Guest OS is independently stored in memory, resulting in multiple occupations of the memory space. In addition, since there may be duplicate parts in the container images of each VM, these images are also loaded separately in memory, further increasing the memory usage. Additionally, when multiple VMs are deployed or upgraded simultaneously, they will download the required images from the container image repository at the same time. This leads to the repeated download of the same files, thus prolonging the deployment and upgrade time.

[0080] Based on this, the embodiments of this application provide a resource management method. A part of the memory in the physical machine's memory space is divided as a shared memory space (or shared memory disk), and the shared memory disk stores the read-only data that can be shared among multiple VMs running on the HOST. After creating the shared memory disk, technologies such as VirtIO-FS are used to mount the shared memory disk to the target VM, enabling the target VM to access the read-only data in the shared memory disk.

[0081] Please refer to Figure 3 , Figure 3 which is the schematic diagram of the process of creating and mounting the shared memory disk in the resource management method provided by the embodiments of this application.

[0082] 301. The management node sends a request to create a shared memory disk to the NFVI node;

[0083] The management node instructs the NFVI node to create a shared memory disk for storing the read-only part of the Guest OS and / or the read-only part of the container image. In step 301, the specific process of the management node sending a creation request to the NFVI node. In one possible implementation, the NFVO node sends the creation request to the VIM node, and then the VIM node forwards the creation request to the NFVI node.

[0084] Specifically, there are two ways for the NFVO node to send a creation request to the VIM node: First, based on the NFV lifecycle management process, the NFVO node sends a creation request to the VNFM node according to the Or-Vnfm interface, and then the VNFM node forwards this creation request to the VIM node according to the Vi-Vnfm interface; Second, directly manage virtualized resources during creation, and the NFVO node directly sends this creation request to the VIM node according to the Or-Vi interface. After receiving this creation request, the VIM node forwards this creation request to the NFVI node according to the Nf-Vi interface. Specific details are not limited here.

[0085] Specifically, the creation request carries:

[0086] 1) Shared memory disk type: including disk, tmpfs, etc.; among them, the disk-type memory disk is a memory mapping based on a disk device, which maps the space on the disk device to memory and uses the interface of the disk device to access the data in memory; the tmpfs-type memory disk is a memory-based file system that uses the memory in the system to create a file system and mounts this file system to the system.

[0087] 2) Path of the shared memory disk on the physical machine: To locate and access the shared memory disk, it is necessary to provide its storage location on the underlying physical machine. This usually involves the path of the file system to facilitate the VNFM and NFVO to correctly mount and manage this disk.

[0088] 3) Size of the occupied memory: This parameter specifies the amount of memory occupied by the shared memory disk. Determine an appropriate size according to the size of the Guest OS and container image and the expected workload to avoid resource bottlenecks or overuse of memory.

[0089] 4) Identifier of the shared memory disk: Create a unique identifier to uniquely identify this shared memory disk within the system. It helps subsequent management and configuration operations to ensure that the correct resources are operated on.

[0090] 5) Whether NUMA (Non-Uniform Memory Access) affinity: NUMA is a memory management strategy where each processor has its own local memory access. If NUMA affinity is selected, the shared memory disk will be placed on a specific NUMA node, which helps to optimize performance.

[0091] 6) numa id: When NUMA affinity is selected, specify the corresponding NUMA node ID. This ensures that the shared memory disk is placed in an appropriate physical memory area so that the processor can access it more efficiently.

[0092] In a possible implementation, default values can be set for Parameter 1), Parameter 2), and Parameter 3). In addition, in addition to the above parameters, other relevant configuration information may be included, such as permission settings, security policies, etc., to ensure the correct configuration and use of the shared memory disk, which is not specifically limited here.

[0093] 302. The NFVI node creates a shared memory disk;

[0094] Specifically, the NFVI node parses the creation request and creates a directory for the shared memory disk according to the user-specified path or default path of Parameter 2); then, it processes according to the memory disk type indicated by Parameter 1). For example, if it is of the tmpfs type, the tmpfs file system is mounted to the newly created directory, and the maximum memory occupancy of the directory is configured according to Parameter 3). Further, if MUNA affinity is selected in Parameter 5), the numa id specified in Parameter 6) is added to the mount parameters to specify the numa id corresponding to tmpfs. Record the correspondence between the created shared memory disk directory and the identifier specified in Parameter 4)

[0095] In addition, in the process of creating a shared memory disk, related operations such as verification tests, error handling, and logging are also included, and the specific details are not elaborated here one by one.

[0096] In this application, in the foregoing NFV system architecture, the role of the NFVI node is to provide a virtualized environment so that network functions can run on general-purpose hardware. The NFVI node divides a part of the virtualized storage resources into memory resources as a shared memory disk for storing the shareable data of multiple VMs running on the NFVI node.

[0097] 303. The management node sends a mount request for the shared memory disk to the NFVI node;

[0098] After the shared memory disk is created, during the process of creating a VM for the VNF virtual network function or after creating the VM, the management node sends a request to mount the shared memory disk to the NFVI node.

[0099] In a possible implementation, the NFVO node sends a mount request for the shared memory disk to the VNFM node according to the Or-Vnfm interface, and then the VNFM node forwards the mount request to the VIM node according to the Vi-Vnfm interface. The VIM node forwards the mount request to the NFVI node according to the Nf-Vi interface.

[0100] Specifically, the mount request carries:

[0101] 1) Shared memory disk identifier: The identifier of the shared memory disk created in the foregoing steps 301-302.

[0102] It can be understood that in practical applications, the NFVI node can create multiple shared memory disks. For example, memory disk 1 and memory disk 2 are created. Among them, memory disk 1 is used to store the read-only data of the operating systems of multiple VMs, while memory disk 2 is used to store the read-only data of the container images used by multiple VMs. Or more memory disks can be created to distinguish different types of container images. When mounting and using, different memory disks are distinguished by the identifiers of the shared memory disks. It can improve the flexibility of data management and enhance the security of data.

[0103] 2) Sharing method: including VirtIO-FS, nfs, and 9pfs, etc. This parameter determines the sharing mechanism used by the shared memory disk. Different sharing methods have their own advantages, disadvantages, and application scenarios. For example, VirtIO-FS is a fast and efficient sharing mechanism based on the Linux kernel and is suitable for high-performance computing scenarios; nfs and 9pfs are network-based file system protocols and are suitable for scenarios where files or directories need to be shared across hosts. If the sharing method is not specified, the NFVO node can provide a default value for it for emergencies.

[0104] 3) Unique identifier of the VM: used to identify the VM instance that needs to use the shared memory disk. During the creation process of the VM, this identifier is not required because resources have not been allocated to the VM at this time. But after the VM is created, in order to associate the shared memory disk with a specific VM instance, this identifier needs to be provided. Through this identifier, the NFVO node can ensure that each VM instance can access the correct shared memory disk.

[0105] 304. The NFVI node mounts the shared memory disk to the target VM.

[0106] The NFVI node parses the mount request, queries the corresponding directory according to the identifier of the shared memory disk in the mount request, and then mounts the shared memory disk to the target VM. The target VM is the newly created VM or the VM that uses the VM identifier carried in the request.

[0107] Exemplarily, if it is mounted during VM creation, when creating the VM, the shared memory disk is mounted for the virtual machine with the specified sharing method and the identifier of the shared memory disk.

[0108] Exemplarily, if it is after the VM is created, according to the unique identifier of the VM, the shared memory disk is mounted for the specified VM with the specified sharing method and the identifier of the shared memory disk.

[0109] It can be understood that based on steps 303 - 304, one or more shared memory disks can be mounted for any VM running on the NFVI.

[0110] It should be noted that the aforementioned interfaces such as Or-Vnfm, Vi-Vnfm, Or-Vi, and Nf-Vi are all interface names defined by ETSI. Through these interfaces, the interaction between various entity nodes in the NFV system is realized. In this context, the interfaces mentioned in the embodiments of the present application are only one of the multiple possible options and do not constitute a mandatory limitation on the interaction interfaces between entities.

[0111] After the shared memory disk is mounted on the VM, the following describes how to use the shared memory disk to load the Guest OS and container images on the VM. Specifically, the management node in the NFV system instructs the VM to implement the loading of the data in the shared memory disk.

[0112] I. Loading the read-only data of the operating system.

[0113] Please refer to Figure 4 , Figure 4 which is a schematic flowchart of loading the Guest OS according to the shared memory disk provided by the embodiments of the present application. At this time, the management node that controls the loading of the Guest OS is the VNFM node.

[0114] 401. The VNFM node sends a request to start the VM to the VM;

[0115] After the NFVI node mounts the shared memory disk for the target VM, the VNFM node sends a configuration file for starting the VM to the VM. In one possible implementation, the information in the configuration file includes:

[0116] 1) The identifier of the shared memory disk: The VM may mount multiple shared memory disks, and it is necessary to instruct the VM to use the shared memory disk storing the Guest OS for processing.

[0117] 2) Whether to use the shared memory disk to share the Guest OS: Some VMs may need to use the shared memory disk to share their Guest OS in order to share computing resources and data between VMs. If the configuration file instructs the VM to use the shared memory disk to share the Guest OS, the VM will use this shared memory disk to load and run the Guest OS.

[0118] In addition, in addition to the above parameters, other relevant configuration information may also be included, such as startup parameters (memory size, number of CPUs, etc.), network configuration (IP address, subnet mask, default gateway, etc.), security configuration, etc., which are set according to specific business requirements and are not specifically limited here.

[0119] In a possible implementation, the specific implementation of step 401 is as follows: The VNFM node sends a request to start a VM to the VIM node according to the Vi-Vnfm interface, and the request carries: the configuration file for starting the VM. The VIM node forwards the request to the NFVI node according to the Nf-Vi interface, and the NFVI node injects the configuration file into the target VM indicated in the request.

[0120] 402. The VM combines the read-only data and read-write data of the Guest OS;

[0121] In a possible implementation, the target VM combines the directory corresponding to the read-only data in the shared memory disk and the directory corresponding to the read-write data in its own memory space through overlayfs, and the Guest OS starts from the combined directory. Among them, overlayfs is a file system service for Linux that can overlay the contents of multiple directories on another directory.

[0122] Please refer to Figure 5 , Figure 5 This is a schematic diagram of sharing container images in the embodiments of the present application. There are 2 virtual machines running in the HOST, namely VM1 and VM2. Among them, the directory name of the first directory combined through overlayfs is / usr, which includes the read-write (rw) data directory work dir of the Guest OS and the shared read-only (ro) data directory share dir. The rw data comes from the memory file system (tmpfs) of the virtual machine, and the ro data comes from the shared memory disk (share tmpfs) of the NFVI.

[0123] In a possible implementation, the rw data includes directories such as the system configuration file directory ( / etc) and the storage change file directory ( / var) of the Guest OS, and the ro data includes directories such as the additional software package directory ( / opt) of the Guest OS. In the present application, the shared ro data is the file directory that multiple VMs in the HOST will not edit when running the Guest OS, and the file directory is not specifically limited.

[0124] 403. The VM starts the Guest OS in the combined directory for virtual machine initialization.

[0125] After injecting the configuration file, the initialization process of the VM starts the Guest OS in the form of the root directory in the combined directory. Specifically, the initialization process of the VM includes the allocation of hardware resources, the startup of the operating system, the configuration of network resources, etc. The resource management method provided by the embodiments of the present application is limited to the startup stage of the operating system of the VM and does not impose any restrictions on its initialization process.

[0126] II. Load the read-only data of the container image.

[0127] As Figure 6 shown, Figure 6 FIG. is a schematic flowchart of running a container image according to a shared memory disk provided by an embodiment of the present application. At this time, the management node that sends a container image running instruction to the VM is the CCM node.

[0128] 601. The CCM node instructs the container cluster service CIS in the VM to use the container image in the shared memory disk;

[0129] After the VM mounts the shared memory disk, the CCM node directly instructs the CIS in the VM to run the container image. In a possible implementation manner, the CCM node receives a container image usage request from the NFVO node, and the request information carries the identifiers of the target VM and the shared memory disk. The CCM node forwards the request to the CISM node, so that the CISM node instructs the CIS in the VM to run the container image.

[0130] 602. The CIS establishes a soft link with the shared memory disk and runs the service corresponding to the container image.

[0131] In a possible implementation manner, a shared memory disk is created on the HOST and mounted to the VM using VirtIO-FS. The container engine directory is divided into two parts: shareable and non-shareable. The shareable part is the container image layer (snapshots), and the non-shareable part is the container metadata (metadata). The non-shareable part is placed in the shared memory disk (share tmpfs), and the shareable part is placed in the shared memory disk. When downloading a container image, first check whether it already exists in the shared memory disk. If it already exists, create a soft link (layer link) in the memory disk and count in the shared memory disk. If it does not exist, download the container image to the shared memory disk, then create a soft link and count.

[0132] Please refer to Figure 7 , Figure 7 FIG. is a schematic diagram of sharing a container image in an embodiment of the present application. There are 2 virtual machines, VM1 and VM2, running in the HOST. There are 3 container images stored in the shared memory disk, including layer1, layer2, and layer3. Among them, VM1 accesses the container images layer1 and layer2 by establishing soft links layer1 link and layer2 link respectively, and VM1 accesses the container images layer2 and layer3 by establishing soft links layer2 link and layer3 link respectively. And count the usage in the shared memory disk. For example, the usage count of layer1 is 1, the usage count of layer2 is 2, and the usage count of layer3 is 1.

[0133] In a possible implementation, according to business adjustment, VM1 needs to run the business carried by container image layer4. If the container image does not exist in the shared memory disk, it is necessary to first download the container image to the shared memory disk. After the download is completed, VM1 accesses the container image layer4 by creating a soft link layer4 link, and at the same time increments the usage count of layer4 in the shared memory disk by 1. If VM2 also needs to run the business of container image layer4, it only needs to create a soft link layer4 link to access the container image layer4 in the shared memory disk and increment the usage count by 1, without repeated downloading.

[0134] Furthermore, if a VM no longer needs to run a certain container image to execute the corresponding business, it is necessary to revoke the previously created soft link and decrement the corresponding usage count by 1. For example, if VM1 no longer needs to run the business corresponding to layer1, cancel the soft link of layer1 link and decrement the usage count corresponding to layer1 by 1. When the usage count reaches 0, it means that the image is no longer used by any VM, and the shared memory disk can safely delete the container image to save storage space.

[0135] In the embodiments of the present application, by managing the infrastructure NFVI nodes in the NFV system, one or more shared memory disks are created in the HOST, which are specifically used to store the read-only data required by multiple VMs. Since each VM runs using the memory resources of the HOST, extracting the read-only data in the memory of multiple VMs and storing them separately can save the memory space required by each VM, reduce the repeated downloading and repeated occupation of memory space by data in different VMs, and improve the utilization rate of memory resources. Further, the saved memory space can also be used to expand more new VMs. In addition, since the read-only data is stored in the shared memory disk, it ensures that the data loading process has high security and consistency, avoiding data conflicts or inconsistencies caused by simultaneous access by multiple users. Further, the shared memory disk can be mounted on the VM, and both the Guest OS and the container image can directly read and run the data from the memory of the HOST, improving the startup and loading speed and enhancing the reliability of the network element with storage bypass.

[0136] Based on the same technical concept, the embodiments of the present application also provide a resource management device for implementing the method embodiments corresponding to the above NFVI nodes. Please refer to Figure 8 , Figure 8A structural schematic diagram of a resource management device provided by an embodiment of the present application. The resource management device 800 includes a receiving module 801 and a processing module 802. The receiving module 801 can implement corresponding communication functions, and the processing module 802 is used for data processing. The receiving module 801 can also be referred to as a communication interface or a communication unit. Among them, both the receiving module 801 and the processing module 802 can be implemented by software or can be implemented by hardware.

[0137] Optionally, the resource management device 800 may further include a storage unit, which can be used to store instructions and / or data. The processing module 802 can read the instructions and / or data in the storage unit to enable the network device to implement the foregoing method embodiments.

[0138] The resource management device 800 can be used to perform the actions in the foregoing method embodiments. The resource management device 800 can be a network device or a component configurable in a network device. The receiving module 801 is used to perform the receiving-related operations in the foregoing method embodiments, and the processing module 802 is used to perform the processing-related operations in the foregoing method embodiments.

[0139] As an example, the resource management device 800 is applied to a network function virtualization NFV system. The NFV system includes a management node and a network function virtualization infrastructure NFVI node. At least two virtual machines VMs are running in the NFVI node; the resource management device 800 is disposed in the NFVI node:

[0140] The receiving module 801 is used to receive a first request from the management node. The first request includes information of a target VM, and the target VM is included in at least two VMs;

[0141] The processing module 802 is used to map the read-only data in the shared memory space to the file path of the target VM according to the information of the target VM. The read-only data is the operating system data and / or container image data of the VM. The operating system data is used to start the target VM, and the container image data is used for the target VM to run the service corresponding to the container image data.

[0142] In addition, an embodiment of the present application further provides another resource management device for implementing the foregoing method embodiments corresponding to the VM. Please refer to Figure 9 , Figure 9 A structural schematic diagram of a resource management device provided by an embodiment of the present application. The resource management device 900 includes a receiving module 901 and a processing module 902. The receiving module 901 can implement corresponding communication functions, and the processing module 902 is used for data processing. The receiving module 901 can also be referred to as a communication interface or a communication unit. Among them, both the receiving module 901 and the processing module 902 can be implemented by software or can be implemented by hardware.

[0143] Optionally, the resource management device 900 may further include a storage unit, which may be used to store instructions and / or data. The processing module 902 may read the instructions and / or data in the storage unit to enable the network device to implement the foregoing method embodiments.

[0144] The resource management device 900 may be used to perform the actions in the foregoing method embodiments. The resource management device 900 may be a network device or a component configurable in a network device. The receiving module 901 is used to perform the receiving-related operations in the foregoing method embodiments, and the processing module 902 is used to perform the processing-related operations in the foregoing method embodiments.

[0145] As an example, the resource management device 900 is applied to a network function virtualization (NFV) system. The NFV system includes a network function virtualization infrastructure (NFVI) node, and at least two virtual machines (VMs) are running in the NFVI node. The at least two VMs are each associated with a shared memory space; the resource management device 900 is disposed in a target VM, and the target VM is one of the at least two VMs; wherein:

[0146] The receiving module 901 is used to obtain read-only data in the shared memory space, and the read-only data is operating system data and / or container image data shared by at least two VMs;

[0147] The processing module 902 is used to perform initialization and / or run the services corresponding to the container image data according to the operating system data.

[0148] This application also provides a resource management device 100. As Figure 10 shown, the resource management device 100 includes: a bus 102, a processor 104, a memory 106, and a communication interface 108. The processor 104, the memory 106, and the communication interface 108 communicate with each other through the bus 102. The resource management device 100 may be a server or a terminal device. It should be understood that this application does not limit the number of processors and memories in the resource management device 100.

[0149] The bus 102 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 10It is represented by only one line in the figure, but it does not mean that there is only one bus or one type of bus. The bus 104 may include a path for transmitting information between various components of the resource management device 100 (for example, the memory 106, the processor 104, and the communication interface 108).

[0150] The processor 104 may include any one or more of processors such as a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0151] The memory 106 may include a volatile memory, such as a random access memory (RAM). The processor 104 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0152] The memory 106 stores executable program codes, and the processor 104 executes the executable program codes to respectively implement the functions of the foregoing receiving module, processing module, and sending module, thereby implementing the resource management method. That is to say, the memory 106 stores instructions for executing the resource management method.

[0153] Alternatively, the memory 106 stores executable codes, and the processor 104 executes the executable codes to respectively implement the functions of the foregoing resource management device, thereby implementing the resource management method. That is to say, the memory 106 stores instructions for executing the resource management method.

[0154] The communication interface 103 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the resource management device 100 and other devices or communication networks.

[0155] The embodiment of the present application also provides a computer program product including instructions. The computer program product may be a software or program product including instructions that can run on a computing device or be stored in any available medium. When the computer program product runs on at least one computing device, at least one computing device is caused to execute the resource management method.

[0156] The embodiments of the present application also provide a computer-readable storage medium. The computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center including one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state drive), etc. The computer-readable storage medium includes instructions that direct the computing device to execute the resource management method.

[0157] 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 foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A network function virtualization (NFV) system, characterized in that, Including: A management node and a Network Function Virtualization Infrastructure (NFVI) node, where at least two Virtual Machines (VMs) are running in the NFVI node; The NFVI node is used to receive a first request from the management node, and the first request includes information of a target VM, and the target VM is included in the at least two VMs; The NFVI node is used to map read-only data in a shared memory space to a file path of the target VM according to the information of the target VM, and the read-only data is operating system data and / or container image data shared by the at least two VMs, the operating system data is used to initialize the target VM, and the container image data is used for the target VM to run a service corresponding to the container image data; The target VM is used to initialize according to the operating system data and / or run a service corresponding to the container image data.

2. The system according to claim 1, characterized in that When the read-only data is operating system data shared by the at least two VMs, the management node is a Virtual Network Function Manager (VNFM) node, and the target VM is specifically used for: Merging the shared operating system data with first data to obtain a first directory, where the first data is read-write data of the operating system of the target VM, and the first data is stored in the memory space of the target VM; Initializing based on the first directory.

3. The system according to claim 1 or 2, characterized in that, When the read-only data is container image data shared by the at least two VMs, the management node is a Container Cluster Manager (CCM) node, and the target VM is specifically used for: Running a service corresponding to the container image data based on container metadata and the shared container image data, and the container metadata is stored in the memory space of the target VM.

4. The system according to any one of claims 1-3, characterized in that, The NFVI node is further used for: Receiving a second request from the management node, and the second request includes an identifier of the shared memory space; Establishing an association between the shared memory space and each of the at least two VMs.

5. The system according to any one of claims 1-4, characterized in that, The NFVI node is further used for: Receiving a third request from the management node, where the third request is used to instruct the NFVI node to create the shared memory space, and the third request includes a memory size to be occupied; Determining, from a virtualized memory space, a memory space that meets the occupied memory size as the shared memory space.

6. The system according to claim 5, wherein The management node includes a Network Function Virtualization Orchestrator (NFVO) node, a Virtual Network Function Manager (VNFM) node, and a Virtual Infrastructure Manager (VIM) node; The VIM node is used to send the third request to the NFVI node according to a fourth request; The NFVO node is used to send the fourth request to the VIM node or the VNFM node; The VNFM node is used to forward the fourth request to the VIM node.

7. A resource management method, characterized in that, Applied to a Network Function Virtualization (NFV) system, the NFV system includes a management node and a Network Function Virtualization Infrastructure (NFVI) node, where at least two Virtual Machines (VMs) are running in the NFVI node, and the method includes: The NFVI node receives a first request from the management node, where the first request includes information about a target VM, and the target VM is included in the at least two VMs; The NFVI node maps read-only data in the shared memory space to the file path of the target VM according to the information of the target VM. The read-only data is operating system data and / or container image data shared by the at least two VMs. The operating system data is used to start the target VM, and the container image data is used for the target VM to run the service corresponding to the container image data.

8. The method according to claim 7, characterized in that, The management node is a virtual network function manager (VNFM) node or a container cluster manager (CCM) node.

9. The method according to claim 7 or 8, characterized in that Before the NFVI node receives the first request from the management node, the method further includes: The NFVI node receives a second request from the management node, where the second request includes an identifier of the shared memory space; The NFVI node establishes an association between the shared memory space and each of the at least two VMs according to the identifier of the shared memory space.

10. The method according to claim 9, wherein Before the NFVI node receives the second request from the management node, the method further includes: The NFVI node receives a third request from the management node, where the third request is used to instruct the NFVI node to create the shared memory space, and the third request includes the occupied memory size; The NFVI node determines, according to the third request, a memory space that meets the occupied memory size from the virtualized memory space as the shared memory space.

11. The method according to any one of claims 10, characterized in that, The management node includes a network function virtualization orchestrator (NFVO) node, a virtual network function manager (VNFM) node, and a virtualized infrastructure manager (VIM) node; the third request is generated by the VIM node according to a fourth request, and the fourth request is sent by the NFVO node to the VIM node or the VNFM node forwards the fourth request of the NFVO to the VIM node.

12. A resource management method, characterized in that, Applied to a network function virtualization (NFV) system, the NFV system includes a network function virtualization infrastructure (NFVI) node, and at least two virtual machines (VMs) are running in the NFVI node. Each of the at least two virtual machines (VMs) is associated with a shared memory space. The method includes: The target VM obtains the read-only data in the shared memory space. The read-only data is operating system data and / or container image data shared by the at least two VMs, and the target VM is one of the at least two VMs; The target VM initializes according to the operating system data and / or runs the service corresponding to the container image data.

13. The method according to claim 12, wherein When the read-only data is the operating system data shared by the at least two VMs, the target VM initializes according to the operating system data, including: The target VM merges the shared operating system data with first data to obtain a first directory. The first data is the read-write data of the operating system of the target VM, and the first data is stored in the memory space of the target VM; The target VM is initialized based on the first directory.

14. The method according to claim 12 or 13, characterized in that, When the read-only data is container image data shared by the at least two VMs, the target VM runs the service corresponding to the container image data, including: The target VM runs the service corresponding to the container image data based on container metadata and the shared container image data, and the container metadata is stored in the memory space of the target VM.

15. A resource management device, characterized in that, Applied to a network function virtualization NFV system, the NFV system includes a management node and a network function virtualization infrastructure NFVI node, and at least two virtual machines VMs are running in the NFVI node; The resource management device is arranged in the NFVI node, and the resource management device includes: A receiving module, configured to receive a first request from the management node, where the first request includes information of a target VM, and the target VM is included in the at least two VMs; A processing module, configured to map the read-only data in the shared memory space to the file path of the target VM according to the information of the target VM, where the read-only data is operating system data and / or container image data of the VM, the operating system data is used to start the target VM, and the container image data is used for the target VM to run the service corresponding to the container image data.

16. A resource management device, characterized in that, Applied to a network function virtualization NFV system, the NFV system includes a network function virtualization infrastructure NFVI node, at least two virtual machines VMs are running in the NFVI node, and the at least two virtual machines VMs are respectively associated with a shared memory space; The resource management device is arranged in the target VM, the target VM is one of the at least two VMs, and the resource management device includes: A receiving module, configured to obtain the read-only data in the shared memory space, where the read-only data is operating system data and / or container image data shared by the at least two VMs; A processing module, configured to perform initialization and / or run the service corresponding to the container image data according to the operating system data.

17. A resource management device, characterized in that, Comprising at least one processor, coupled to a memory; The memory is used to store programs or instructions; The at least one processor is used to execute part or all of the programs or instructions, so that the method according to any one of claims 7 to 11 is executed.

18. A resource management device, characterized in that, Comprising at least one processor, coupled to a memory; The memory is used to store programs or instructions; The at least one processor is used to execute part or all of the programs or instructions, so that the method according to any one of claims 12 to 14 is executed.

19. A computer-readable storage medium, comprising instructions, when the instructions run on a computer, so that the method according to any one of claims 7 to 11 is executed, or, so that the method according to any one of claims 12 to 14 is executed.

20. A computer program product containing instructions, when it runs on a computer, so that the method according to any one of claims 7 to 11 is executed, or, so that the method according to any one of claims 12 to 14 is executed.

Images