Remote trusted operation control method for virtual machine container architecture in cloud storage
By introducing two-layer proxy access and dual-channel wireless communication in cloud storage, dynamic isolation of non-licensed and licensed storage cells is solved, and the remote trusted operation and security of the container architecture of cloud storage virtual machine is ensured. , inspires interdisciplinary talents to innovate.
Patent Information
- Application Number
- CN202210086193.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-01-25
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2042-01-25
AI Technical Summary
The software isolation mechanism of Docker containers makes it weak in isolation, which can easily lead to security issues such as permission escalation, container escape and denial of service, affecting the remote and trustworthy operation of the virtual machine container architecture in cloud storage.
In a system composed of local physical machines, cloud storage virtual machines and clients, through dual-layer proxy access and dual-channel wireless communication, the main body and object metrics form modules are used for authentication, and non-licensed and licensed storage cells are dynamically isolated to ensure that non-licensed and licensed storage cells are ensured. Security and isolation of information transmission paths.
Effectively prevent illegal users from intruding into the container architecture, ensure the remote and trustworthy operation of the virtual machine container architecture in cloud storage, encourage interdisciplinary talents to participate in technological innovation, and realize the security and isolation of subject and object identity verification.
Smart Images

Figure CN114398154B_ABST
Abstract
Description
Technical Field
[0001] The invention discloses a control method for remote trusted operation of a virtual machine container architecture in cloud storage, belonging to the technical field of remote trusted operation of a cloud storage container architecture. Background Art
[0002] As a new type of information technology, cloud computing uses virtualization technology to cleverly combine the acquisition, deployment and combination of services, which is completely changing many IT ecosystems. Container technology is an operating system (OS) level virtualization technology. Each container can be regarded as an independent operating system instance. By isolating the applications in each container and sharing physical service resources, it achieves similar effects to hardware virtualization technology, making the Docker container running on the host operating system have advantages in performance and portability. However, the software isolation mechanism of Docker containers makes its isolation weak, which easily leads to the host system facing privilege escalation (allowing users to obtain privileges they do not originally have), container escape (processes in the container cannot see processes outside, but processes outside can see inside, so if a container can access external resources and even obtain the privileges of the host host, it is called container escape), denial of service (all containers share the same kernel resources. If a set of containers can exclusively access certain resources, including memory and other more abstract resources such as user IDs, then other containers on the same host are likely to be unable to operate normally due to lack of resources. This is a denial of service attack), which leads to a series of security and privacy issues that need to be solved. Virtual machine technology is the core supporting technology of cloud computing, and Docker is the implementation carrier of cloud virtualization technology. Based on trusted computing technology, ensuring the code and data integrity of services running in the Docker environment is an important foundation for building a trusted cloud environment and achieving trusted consensus between cloud service providers and cloud tenants. Summary of the invention
[0003] The purpose of the present invention is to provide a security control method for the remote trusted operation of the virtual machine container architecture in the enterprise's management innovation, technology innovation, talent innovation and cloud storage.
[0004] The present invention is characterized in that it is implemented in the following steps in a local cloud storage virtual machine container architecture remote trusted operation control system composed of a local physical machine, a local cloud storage virtual machine and a client, hereinafter referred to as the system:
[0005] Step (1) system construction, wherein:
[0006] Local physical machine, including: controller CPU S, a subject measurement value forming module, an object measurement value forming module and a local sample database, wherein:
[0007] The local sample database stores local samples that have been included in national standards, enterprise standards, industry standards, or have been widely recognized in the industry but have not yet been included in the standards. Each local sample is represented in the form of a linked list: the subject feature value of the inventor → the subject hash value → the subject measurement value, and the object feature value of the application → the object hash value → the object measurement value.
[0008] Controller CPU S First, the characteristic information of the local inventor is selected as the subject natural person characteristic value and the subject social person characteristic value, and then these two types of subject characteristic values are concatenated into a unified subject hash value, which is sent to the subject measurement value formation module, and a subject measurement value is obtained after trusted calculation. Similarly, the characteristic information of the application program submitted by the local inventor as the object is selected as the object hash value, which is sent to the object measurement value formation module, and an object measurement value is obtained after trusted calculation. The controller CPU S Organize the local inventors into a subject sample represented by the above two linked lists and send it to the local sample database for storage.
[0009] The local cloud storage virtual machine is a container architecture composed of all containers, which is a virtual machine controller CPU V(非特许) 、CPU V(特许) The container architecture of the control includes: the virtual machine controller CPU V , and a worm net that holds all the containers, including:
[0010] The outer contour of the worm network is formed by connecting containers located outside the container architecture. The local containers are referred to as nodes, the containers on the outer contour are called edge nodes, and the containers in the network are called internal nodes. In the network, there are two non-franchised areas and franchised areas whose storage capacity, i.e., the number of nodes, changes dynamically with customer needs, respectively located on the left and right sides of the worm network.
[0011] The virtual machine controller CPU V It is set on the edge node closest to the physical machine on the outer contour line of the worm network, that is, the central control node.
[0012] Each node is provided with a node controller CPU. m , m = 1, 2, …, m, …, M, M is the total number of nodes, m is the node number, the nodes are classified according to the purpose of the application, and store at least one application;
[0013] Step (2) System initialization:
[0014] Step (2.1), physical machine controller CPU S Initialize the array to the main hash value: {user name, user password, user number, user work unit, work permit number},
[0015] Object hash value: {object name, object version number, object publication date, object confidentiality level},
[0016] Central control node to each node control CPU m The URL mapping table is used to determine the shortest, fastest and safest path between two nodes or between any node and the central control node. The calculation steps are as follows:
[0017] Step (2.1.1), assuming the starting node is A, the ending node is B, and a straight line connects A and B.
[0018] Step (2.1.2), draw a perpendicular line along the connecting line between the two nodes, and then send a signal to find a node with the shortest distance to A within the 180-degree space ahead. If it cannot be found, change to a node that is the closest.
[0019] Step (2.1.3), repeat step (2.1.2),
[0020] Step (2.1.4) until the last node found is the terminal node B. Connecting all the nodes in this process is the shortest, fastest and safest path.
[0021] Step (2.2), initialization of the storage area, establishing non-licensed storage cells and licensed storage cells respectively, the steps are as follows: establish non-licensed storage cells on the left, establish licensed storage cells on the right,
[0022] A zero starting point for the storage address is established in the container on the outer contour line at the lower left corner of the storage area, and each container is placed equidistantly along the axis of each container in the direction of increasing storage addresses, and so on from left to right until the right side, forming a container matrix with its upper and lower ends both on the outer contour line.
[0023] The second container located on the outer contour line at the upper right corner of the storage area is established as the starting point of the privileged storage address. The second storage address coordinate system is established in the container matrix in the same way as above along the direction of decreasing storage addresses. Thus, in the same container matrix, two storage address coordinate systems with overlapping spatial positions but completely complementary address coordinate systems are set. From left to right, the storage address coordinate systems with storage address numbers starting from zero and increasing in sequence belong to the non-privileged storage area. Otherwise, they belong to the privileged storage area. Under normal operating conditions, the sum of the two capacities must be less than the sum of the capacities in the storage area. Moreover, the respective capacities of the two are dynamically changing, and the former must be greater than the latter. Therefore, a dynamic isolation zone is naturally formed between the non-privileged storage area and the privileged storage area to prevent mutual interference.
[0024] Step (2.3), on the physical machine CPU S and VM CPU V Each of the two is provided with a wireless transceiver module with a dual antenna, thus forming a dual-channel wireless communication signal path between the two, transmitting the information of the non-licensed storage cell or the licensed storage cell respectively.
[0025] Step (2.4), in the CPU S A mapping table is set up on the virtual machine, and the same application is located in the non-privileged storage area of the virtual machine. CPU S The mapping table of the addresses of the non-privileged storage area is similarly implemented for the virtual machine licensed area after aligning the starting storage addresses of the corresponding storage areas of both parties and then filling the application programs of the physical machine one by one along the virtual machine node.
[0026] Step (3), the system implements the remote trusted operation of the virtual machine container architecture in the cloud storage by sequentially performing the following steps:
[0027] Step (3.1), the physical machine performs the following steps to authenticate the customer:
[0028] Step (3.1.1), physical machine CPU S Verify the identity of the visiting customer:
[0029] Step (3.1.1.1) verifies whether the measurement value of the visiting customer is in the sample database of the inventor of the unit. If so, it proves that the visiting customer is the inventor of the unit and executes step (3.1.2). Otherwise, it verifies that the visiting customer is not the inventor of the unit and executes step (3.1.1.2).
[0030] Step (3.1.1.2) checks the personal and social information of the visiting customer. If the customer is a member of the company, execute step (3.1.2). Otherwise, the customer is denied access.
[0031] Step (3.1.2), physical machine CPU S Check the object of the visiting customer. If the object of the visiting customer is included in the measurement system of the local sample database, the object identity authentication passes. Otherwise, access is denied.
[0032] The advantages of the present invention are:
[0033] 1. Taking the inventor as the subject, establish the subject's social characteristic information item in the subject's characteristic information. While excluding non-unit personnel from entering, allow the unit's personnel to have a certain social identity, encourage cross-disciplinary and cross-field talents to participate in technological innovation activities, and serve as sample information for the unit's personnel to stimulate their desire for innovation.
[0034] 2. Separate the access and storage functions of the application as the object from the physical machine to the virtual machine of the cloud storage, so that the physical machine can perform the principal and object identity authentication while serving the proxy access function of the access client to isolate the visitor directly outside the entity and prevent him from directly invading the operating system of each container in the container architecture.
[0035] 3. Starting from the signal output end of the physical machine to the container architecture of the virtual machine, the following technologies are adopted to prevent illegal users from using intermediate containers to interfere with each other between non-licensed storage cells and licensed cells:
[0036] (1) On the physical machine CPU S and VM CPU V Two-way dual-channel wireless communication is achieved between them. S and VM CPU V The wireless transceiver module inside achieves isolation between non-licensed and licensed area channels.
[0037] (2) Virtual machine CPU on the central control node V In the worm network, the CPU is connected to the non-privileged storage area and the licensed storage area through different shortest, fastest and safest paths. V The two exit ports reach their respective destination nodes CPU m Thus, the isolation of each information transmission channel is achieved.
[0038] 4. Each node in the storage architecture, including the edge node, has a physical machine CPU. S By CPU V Sent to the node to determine the CPU from any node V The shortest, fastest and safest transmission path between nodes or between any nodes is guaranteed not only from the CPU VThe transmission speed to the node at the non-privileged storage cell storage address starting point at the lower left corner of the worm network outer contour or the privileged storage cell storage address starting point at the upper right corner also ensures the isolation of the non-privileged transmission path and the privileged transmission path in storage space. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] Figure 1 System main program flow chart.
[0040] Figure 2 System hardware block diagram:
[0041] Figure 2.1 System Block Diagram,
[0042] Figure 2.2 Diagram of container storage structure.
[0043] Figure 3 Physical machine block diagram.
[0044] Figure 4 Schematic diagram of the method for determining the shortest, fastest and safest transmission path in a worm network. DETAILED DESCRIPTION
[0045] The present invention adopts a double-layer proxy access method, even if the physical machine CPU S It is also necessary to go through the virtual machine CPUv (non-privileged) and CPUv (privileged) to address and access the non-privileged area and the privileged storage area. This makes it impossible for customers to directly access the above two storage areas through the physical machine CPUs, nor can they directly address and access the virtual machine, which greatly enhances the function of proxy access.
[0046] The present invention is in the physical machine CPU S , virtual machine CPU V(非特许) or CPU V(特许) Bidirectional dual-channel wireless communication, CPU V(非特许) A hub-and-spoke network between nodes in non-licensed storage communities. CPU V(特许) The hub-and-spoke network between the nodes in the licensed cell uses the non-licensed storage cell controller CPU V(非特许) , Licensed storage cell controller CPU V(特许) , realizing the slave CPU in each cell V(非特许) or CPU V(特许) The shortest path between each node is realized from the physical machine CPU S Departure through the CPU V(非特许) 、CPU V(特许) The isolation of the communication paths to the self-accessed addressing terminal nodes prevents the path interference between the non-authorized storage cells and the authorized storage cells, as well as the artificial path interference between the two cells.
[0047] The CPU is set on the access addressing node m , m is the access node number, m = 1, 2, ..., m, ..., M, there is a CPU m →The mapping table of the storage address of the data file in the container→the name of the data file stores the data file in a left-right symmetrical manner along the longitudinal axis of the container, thus achieving fixed-point storage while achieving fast transmission.
[0048] The present invention establishes a set of storage address coordinate systems in the same physical storage space, with the two diagonal edge nodes located at the lower left and upper right of the outer contour line as the starting point and the end point respectively, and establishes two storage areas, non-franchised and franchised, respectively, in a manner in which the address at the starting point is increased and the address at the end point is decreased. As long as the storage area does not overflow, there must be an isolation zone between the two, and the storage capacity of the two increases and decreases, thereby achieving the purpose of dynamic elastic isolation and full utilization of storage space.
Claims
1. A method for controlling remote trusted operation of a virtual machine container architecture in cloud storage, characterized in that: It is a remote trusted operation control system in a local cloud storage virtual machine container architecture composed of a local physical machine, a local cloud storage virtual machine and a client. The following steps are implemented in the system as follows: Step (1) system construction, wherein: Local physical machine, including: controller CPU S , a subject measurement value forming module, an object measurement value forming module and a local sample database, wherein: The local sample database stores local samples that have been included in national standards, enterprise standards, industry standards, or have been widely recognized in the industry but have not yet been included in the standards. Each local sample is represented in the form of a linked list: the subject feature value of the inventor → the subject hash value → the subject measurement value, and the object feature value of the application → the object hash value → the object measurement value. Controller CPU S First, the characteristic information of the local inventor is selected as the subject natural person characteristic value and the subject social person characteristic value, and then these two types of subject characteristic values are concatenated into a unified subject hash value, which is sent to the subject measurement value formation module, and a subject measurement value is obtained after trusted calculation. Similarly, the characteristic information of the application program submitted by the local inventor as the object is selected as the object hash value, which is sent to the object measurement value formation module, and an object measurement value is obtained after trusted calculation. The controller CPU S Organize the local inventors into a subject sample represented by the above two linked lists and send it to the local sample database for storage. The local cloud storage virtual machine is a container architecture composed of all containers, which is a virtual machine controller CPU V(非特许) 、CPU V(特许) The container architecture of the control includes: the virtual machine controller CPU V , and a worm net that holds all the containers, including: The outer contour of the worm network is formed by connecting containers located outside the container architecture. The local containers are referred to as nodes, the containers on the outer contour are called edge nodes, and the containers in the network are called internal nodes. In the network, there are two non-franchised areas and franchised areas whose storage capacity, i.e., the number of nodes, changes dynamically with customer needs, respectively located on the left and right sides of the worm network. The virtual machine controller CPU V It is set on the edge node closest to the physical machine on the outer contour line of the worm network, that is, the central control node. Each node is provided with a node controller CPU. m , m = 1, 2, …, m, …, M, M is the total number of nodes, m is the node number, the nodes are classified according to the purpose of the application, and store at least one application; Step (2) System initialization: Step (2.1), physical machine controller CPU S Initialize the array to the main hash value: {user name, user password, user number, user work unit, work permit number}, Object hash value: {object name, object version number, object publication date, object confidentiality level}, Central control node to each node control CPU m The URL mapping table is used to determine the shortest, fastest and safest path between two nodes or between any node and the central control node. The calculation steps are as follows: Step (2.1.1), assuming the starting node is A, the ending node is B, and a straight line connects A and B. Step (2.1.2), draw a perpendicular line along the connecting line between the two nodes, and then send a signal to find a node with the shortest distance to A within the 180-degree space ahead. If it cannot be found, change to a node that is the closest. Step (2.1.3), repeat step (2.1.2), Step (2.1.4) until the last node found is the terminal node B. Connecting all the nodes in this process is the shortest, fastest and safest path. Step (2.2), initialization of the storage area, establishing non-licensed storage cells and licensed storage cells respectively, the steps are as follows: establish non-licensed storage cells on the left, establish licensed storage cells on the right, A zero starting point for the storage address is established in the container on the outer contour line at the lower left corner of the storage area, and each container is placed equidistantly along the axis of each container in the direction of increasing storage addresses, and so on from left to right until the right side, forming a container matrix with its upper and lower ends both on the outer contour line. The second container located on the outer contour line at the upper right corner of the storage area is established as the starting point of the privileged storage address. The second storage address coordinate system is established in the container matrix in the same way as above along the direction of decreasing storage addresses. Thus, in the same container matrix, two storage address coordinate systems with overlapping spatial positions but completely complementary address coordinate systems are set. From left to right, the storage address coordinate systems with storage address numbers starting from zero and increasing in sequence belong to the non-privileged storage area. Otherwise, they belong to the privileged storage area. Under normal operating conditions, the sum of the two capacities must be less than the sum of the capacities in the storage area. Moreover, the respective capacities of the two are dynamically changing, and the former must be greater than the latter. Therefore, a dynamic isolation zone is naturally formed between the non-privileged storage area and the privileged storage area to prevent mutual interference. Step (2.3), on the physical machine CPU S and VM CPU V Each of the two is provided with a wireless transceiver module with a dual antenna, thus forming a dual-channel wireless communication signal path between the two, transmitting the information of the non-licensed storage cell or the licensed storage cell respectively. Step (2.4), in the CPU S A mapping table is set up on the virtual machine, and the address of the same application in the non-privileged storage area of the virtual machine ←—→ CPU S The mapping table of the addresses of the non-privileged storage area is similarly implemented for the virtual machine licensed area after aligning the starting storage addresses of the corresponding storage areas of both parties and then filling the application programs of the physical machine one by one along the virtual machine node. Step (3), the system implements the remote trusted operation of the virtual machine container architecture in the cloud storage by sequentially performing the following steps: Step (3.1), the physical machine performs the following steps to authenticate the customer: Step (3.1.1), physical machine CPU S Verify the identity of the visiting customer: Step (3.1.1.1) verifies whether the measurement value of the visiting customer is in the sample database of the inventor of this unit. If it exists, it proves that the visiting customer is the inventor of this unit and executes step (3.1.2). Otherwise, it verifies that the visiting customer is not the inventor of this unit and executes step (3.1.1.2). Step (3.1.1.2) checks the personal and social information of the visiting customer. If the customer is a member of the company, execute step (3.1.2). Otherwise, the customer is denied access. Step (3.1.2), physical machine CPU S Check the object of the visiting customer. If the object of the visiting customer is included in the measurement system of the local sample database, the object identity authentication passes. Otherwise, access is denied.
