A load balancing system and method suitable for cloud server cryptographic machine

Abstract

The present invention discloses a load balancing system and method applicable to a cloud server cryptographic machine, and relates to the technical field of cloud computing. The system includes a scheduling module and a cluster module; the scheduling module is connected to the cluster module; the scheduling module detects the cloud server cryptographic machine, obtains virtual cryptographic machine information, and schedules the virtual cryptographic machine virtualized by the cloud server cryptographic machine to build a virtual cryptographic machine cluster according to the virtual cryptographic machine information; the cluster module acts as an agent for the virtual cryptographic machine cluster to provide cryptographic computing services, and the scheduling module notifies the cluster module to dynamically adjust the virtual cryptographic machine cluster according to the virtual cryptographic machine information. The present invention can discover virtual cryptographic machines in a local area network, automatically form a virtual cryptographic machine cluster, and perform load balancing, and maximize the utilization of cloud server cryptographic machine resources by increasing or decreasing the number of virtual cryptographic machine nodes.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing technology, and more specifically to a load balancing system and method suitable for a cloud server cryptographic machine. Background Art

[0002] The cloud server cryptographic machine can virtualize multiple virtual cryptographic machines, each of which is independent of each other and has an independent IP. There are peak periods and off-peak periods when the virtual cryptographic machine is used. During the peak period, multiple virtual cryptographic machines are required to provide cryptographic operations, and multiple virtual cryptographic machines provide cryptographic operation services to the outside world through one IP address; during the off-peak period, only a single virtual cryptographic machine is required to provide cryptographic operations. When multiple virtual cryptographic machines provide cryptographic operation services to the outside world at the same time, there are problems that multiple virtual cryptographic machines cannot automatically cluster components and cannot provide cryptographic operation services to the outside world through one IP address. At the same time, it is also impossible to maximize the utilization of cloud server cryptographic machine resources by dynamically increasing or decreasing the number of virtual cryptographic machines.

[0003] Therefore, how to achieve load balancing of a virtual cryptographic machine cluster and improve resource utilization is an urgent problem that those skilled in the art need to solve. Summary of the invention

[0004] In view of this, the present invention provides a load balancing system and method suitable for cloud server cryptographic machines, which can discover virtual cryptographic machines in a local area network, automatically form a virtual cryptographic machine cluster, and perform load balancing, so as to maximize the utilization of cloud server cryptographic machine resources by increasing or decreasing the number of virtual cryptographic machine nodes.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] A load balancing system for a cloud server cryptographic machine comprises a scheduling module and a cluster module; the scheduling module is connected to the cluster module; the scheduling module detects a cloud server cryptographic machine in a local area network, obtains virtual cryptographic machine information, schedules a virtual cryptographic machine virtualized by the cloud server cryptographic machine to build a virtual cryptographic machine cluster according to the virtual cryptographic machine information and received configuration information, and transmits the virtual cryptographic machine cluster to the cluster module; the cluster module acts as an agent for the virtual cryptographic machine cluster to provide cryptographic operation services, and the scheduling module notifies the cluster module to dynamically adjust the virtual cryptographic machine cluster according to the virtual cryptographic machine information and the configuration information;

[0007] The scheduling module performs fault detection on the virtual cipher machine according to the virtual cipher machine information, and performs stress detection on the virtual cipher machine in combination with the configuration information, adds or removes the virtual cipher machine in the virtual cipher machine cluster according to the fault detection result and the stress detection result, and adjusts the weight of the virtual cipher machine cluster according to the stress detection result.

[0008] Preferably, the cloud server cryptographic machine includes a device management module, which creates several virtual cryptographic machines and assigns a unique cryptographic machine IP to each of the virtual cryptographic machines. At the same time, it monitors the number of connections, service operation status and cryptographic computing power utilization of all the virtual cryptographic machines, and generates virtual cryptographic machine information for transmission to the scheduling module.

[0009] Preferably, the virtual cryptographic machine includes functions such as key management and cryptographic operations, and key management includes functions such as key generation, storage, deletion, and export.

[0010] Preferably, the scheduling module detects several of the cloud server cryptographic machines in the local area network through UDP broadcast, records the virtual cryptographic machine information of all created virtual cryptographic machines to build a virtual cryptographic machine pool, and marks the call status of all the virtual cryptographic machines in the virtual cryptographic machine pool as available.

[0011] Preferably, the scheduling module sends a monitoring reporting address to the device management module, and the device management module sends a monitoring information service address to the scheduling module, thereby realizing connection and communication between the scheduling module and the device management module.

[0012] Preferably, when the cluster module is initialized, the scheduling module selects several virtual cryptographic machines from at least two cloud server cryptographic machines from the virtual cryptographic machine pool as cryptographic machine nodes to build a virtual cryptographic machine cluster; the number of the cryptographic machine nodes is a preset node threshold.

[0013] Preferably, the cluster module records the weight and call status of each of the cryptographic machine nodes, acts as an agent for each of the cryptographic machine nodes, provides cryptographic operation services through a set service IP, and sends service requests to the corresponding cryptographic machine nodes for cryptographic operations according to the cryptographic machine IP.

[0014] Preferably, the cluster module sends a heartbeat check to all the cryptographic machine nodes to obtain a check result. If the check result is a fault type, the check result is fed back to the scheduling module; the scheduling module traverses the virtual cryptographic machine information and generates the check result when a faulty virtual cryptographic machine is found; the scheduling module notifies the cluster module according to the check result to remove the corresponding cryptographic machine node to the virtual cryptographic machine pool and mark it as unavailable, and at the same time retrieves a corresponding number of available virtual cryptographic machines from the virtual cryptographic machine pool and adds them to the virtual cryptographic machine cluster.

[0015] Preferably, the scheduling module continuously receives the virtual cryptographic machine information, and when the service operation status of the virtual cryptographic machine in the unavailable state is normal, the calling status of the corresponding virtual cryptographic machine in the virtual cryptographic machine pool is modified to an available status; the virtual cryptographic machine in the unavailable state cannot be allocated to the virtual cryptographic machine cluster.

[0016] Preferably, the scheduling module calculates the node pressure of each cipher machine node according to the virtual cipher machine information, and adds or removes the cipher machine node according to the comparison result between the node pressure and the set pressure threshold; the node pressure P is expressed as:

[0017] P = Service connection utilization rate N+ cryptographic computing power utilization M

[0018] Among them, the service connection utilization rate is the ratio of the number of connections of the cryptographic machine node to the preset maximum number of connections; N represents the service connection utilization coefficient; M represents the cryptographic computing power utilization coefficient; N+M=1;

[0019] If the node pressure is greater than or equal to the pressure threshold, the scheduling module retrieves the newly available virtual cryptographic machine from the virtual cryptographic machine pool and adds it to the virtual cryptographic machine cluster; if the node pressure is 0 and the number of the cryptographic machine nodes is greater than the preset node threshold, the cryptographic machine node with the node pressure of 0 is removed from the virtual cryptographic machine pool; otherwise, the cryptographic machine node is not adjusted.

[0020] Preferably, the system also includes an operation interface, which receives configuration information including node thresholds, service connection utilization coefficients, and cryptographic computing power utilization coefficients, and the scheduling module dynamically adjusts the number of cryptographic machine nodes according to the configuration information.

[0021] Preferably, when the cluster module and the scheduling module are deployed independently, the scheduling module sends a network request to the cluster module through the configured cluster module IP address, and the network request includes adding or removing the cryptographic machine node and configuring the load balancing mode. The cluster module adds or removes the cryptographic machine node according to the load balancing mode, and the load balancing mode includes a polling mode or a weighted polling mode, etc.

[0022] Preferably, the scheduling module compares the node pressure of each of the cryptographic machine nodes and increases the weight of the cryptographic machine nodes with less node pressure. Dynamically modify the weight of the cluster nodes, increase the weight of the nodes with less pressure P, and the cloud server cryptographic machine with a high weight value will process more requests, dynamically balancing the direct pressure of each cryptographic machine node.

[0023] Preferably, for n+1 of the cryptographic machine nodes, the node pressures are P1, P2, ..., Pn respectively. If , i=1,2,...,n+1, Pi represents the node pressure of the i-th cipher machine node, then the weight of the i-th cipher machine node is increased, otherwise if it is greater than 0, the weight is reduced.

[0024] A load balancing method applicable to a cloud server cryptographic machine comprises the following steps:

[0025] Step 1: The cloud server cryptographic machine is initialized, and the device management module creates several virtual cryptographic machines;

[0026] Step 2: Obtain the virtual cipher machine information of each virtual cipher machine and build a virtual cipher machine pool;

[0027] Step 3: receiving configuration information, and scheduling the virtual cryptographic machine from the virtual cryptographic machine pool as a cryptographic machine node according to the configuration information to build a virtual cryptographic machine cluster;

[0028] Step 4: Dynamically adjust the cryptographic machine nodes and quantity in the virtual cryptographic machine cluster according to the virtual cryptographic machine information and the configuration information.

[0029] It can be seen from the above technical solution that compared with the prior art, the present invention discloses a load balancing system and method suitable for cloud server cryptographic machines, which can intelligently discover virtual cryptographic machines in a local area network, automatically form a virtual cryptographic machine cluster, proxy virtual cryptographic machine nodes in the cluster, provide cryptographic operation services to the outside world through an IP, support load balancing modes such as polling and weighted polling in the cluster, and maximize the utilization of cloud server cryptographic machine resources by increasing or decreasing the number of virtual cryptographic machine nodes. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0031] Figure 1 A schematic diagram of the structure of a load balancing system applicable to a cloud server cryptographic machine in an embodiment provided by the present invention;

[0032] Figure 2 A schematic diagram of a load balancing method for a cloud server cryptographic machine provided by the present invention;

[0033] Figure 3A flow chart of a load balancing method applicable to a cloud server cryptographic machine in an embodiment provided by the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] The embodiment of the present invention discloses a load balancing system suitable for a cloud server cryptographic machine, comprising a scheduling module and a cluster module; the scheduling module is connected to the cluster module; the scheduling module detects the cloud server cryptographic machine in a local area network, obtains virtual cryptographic machine information, schedules the virtual cryptographic machine virtualized by the cloud server cryptographic machine according to the virtual cryptographic machine information and the received configuration information to build a virtual cryptographic machine cluster, and transmits the virtual cryptographic machine cluster to the cluster module; the cluster module acts as an agent for the virtual cryptographic machine cluster to provide cryptographic operation services, and the scheduling module notifies the cluster module to dynamically adjust the virtual cryptographic machine cluster according to the virtual cryptographic machine information and the configuration information; the scheduling module performs fault detection on the virtual cryptographic machine according to the virtual cryptographic machine information, and performs stress detection on the virtual cryptographic machine in combination with the configuration information, adds or removes the virtual cryptographic machine in the virtual cryptographic machine cluster according to the fault detection result and the stress detection result, and also adjusts the weight of the virtual cryptographic machine cluster according to the stress detection result.

[0036] Furthermore, the cloud server cryptographic machine includes a cloud server cryptographic machine host, and the cloud server cryptographic machine host includes a device management module. The device management module is responsible for creating and monitoring virtual cryptographic machines and allocating independent IPs. Each virtual cryptographic machine is independent of each other; the device management module monitors the number of connections, service operation status, and cryptographic computing power utilization of the virtual cryptographic machines to generate virtual cryptographic machine information and report it to the scheduling module of the load balancing system; the virtual cryptographic machine includes functions such as key management and cryptographic operations, and key management includes functions such as key generation, storage, deletion, and export.

[0037] Furthermore, the scheduling module detects multiple groups of virtual cipher machines and cloud server cipher machine hosts in the local area network through UDP broadcast, collects the virtual cipher machine information reported by the device management module of the cloud server cipher machine host, and dynamically increases or decreases the number of cipher machine nodes of the cluster module; the scheduling module records the virtual cipher machine information as a virtual cipher machine pool, and marks the call status of the virtual cipher machine with normal operating status as available and reallocatable; the monitoring reporting address of the scheduling module is sent to the device management module, and the device management module uploads the service address of the monitoring information to the scheduling module, thereby realizing two-way communication.

[0038] Furthermore, when the cluster module is initialized, the scheduling module selects n+1 virtual cryptographic machines as cryptographic machine nodes to build a virtual cryptographic machine cluster, where n≥2, and the selected cryptographic machine nodes span at least two cloud server cryptographic machine hosts, to avoid simultaneous failure of virtual cryptographic machine nodes caused by failure of cloud server cryptographic machine hosts.

[0039] Furthermore, the cluster module acts as an agent for the cipher machine nodes in the virtual cipher machine cluster, records the weights, call status and other information of the virtual cipher machine nodes in the cluster, provides a unified cryptographic operation service portal to the outside world, provides cryptographic operation services to the outside world through an IP address, and sends service requests to the corresponding cipher machine nodes for cryptographic operations according to the IP allocated to the virtual cipher machine; the cluster module supports independent deployment, and can also be deployed on the same device as the scheduling module; the cluster module sends heartbeat detection to the cipher machine nodes, and feeds back to the scheduling module when a virtual cipher machine failure is detected.

[0040] Furthermore, the scheduling module and cluster module can be deployed independently. One scheduling module corresponds to multiple cluster modules. By increasing or decreasing the number of cipher machine nodes under the cluster module, the cipher machine resources of the cloud server can be maximized.

[0041] Furthermore, the scheduling module receives the data reported by the device management module for analysis, and increases or decreases the number of cipher nodes of the cluster module and the weight of the cipher nodes, including:

[0042] (1) The scheduling module receives node failure information from the cluster module, notifies the cluster module to remove the node, and selects an unassigned virtual cipher machine to add a cipher machine node to the virtual cipher machine cluster; the scheduling module traverses the virtual cipher machine information reported by the device management module, and when a faulty virtual cipher machine is found, notifies the cluster module to remove the faulty cipher machine node corresponding to the faulty virtual cipher machine, and adds a node to the cluster; the scheduling module recycles the removed node to the virtual cipher machine pool, marks it as unavailable, and the node is no longer allocated;

[0043] (2) The scheduling module determines if the node pressure P of each cipher node = service connection utilization rate N+ cryptographic computing power utilization If M≥50%, N+M=1, service connection utilization rate=number of connections / preset maximum number of connections, N represents the service connection utilization rate coefficient, and M represents the cryptographic computing power utilization rate coefficient, the cluster is expanded and a new cryptographic machine node is added to the virtual cryptographic machine cluster; otherwise, if the node pressure P=0 and the number of cryptographic machine nodes in the virtual cryptographic machine cluster is greater than n+1, the cluster is reduced and the cryptographic machine nodes with node pressure P=0 are removed; the scheduling module recycles the removed nodes to the virtual cryptographic machine pool, and the nodes can be reallocated;

[0044] (3) Users can directly operate the scheduling module through the operation interface to configure the values ​​of n, M, and N. If the cluster module is deployed independently, configure the cluster module IP address in the scheduling module. The scheduling module sends a network request to the cluster module according to the cluster module IP address, controls the increase or decrease of cipher machine nodes and configures the load balancing mode of the cluster module according to the network request, and the cluster module increases or decreases cipher machine nodes according to the load balancing mode. The load balancing mode includes polling, weighted polling, etc.; it supports automatic cluster configuration, and also supports users to manually configure cluster parameters, including weights, etc.

[0045] (4) The scheduling module compares the node pressure P of each cipher machine node and dynamically modifies the weight of the cipher machine nodes in the virtual cipher machine cluster, increasing the weight of the node with a small node pressure P. The server with a high weight value will process more requests and dynamically balance the direct pressure of each node. For the n+1 cipher machine nodes, the node pressures are P1, P2, ..., Pn respectively. If , i=1,2,...,n+1, Pi represents the node pressure of the i-th cipher machine node, then the weight of the i-th cipher machine node is increased, otherwise if it is greater than 0, the weight is reduced.

[0046] On the other hand, a load balancing method suitable for cloud server cryptographic machines, such as Figure 2 As shown, the following steps are included:

[0047] Step 1: The cloud server cryptographic machine is initialized, the device management module creates several virtual cryptographic machines, and assigns a unique IP to each virtual cryptographic machine;

[0048] Step 2: Obtain the virtual cipher machine information of each virtual cipher machine and build a virtual cipher machine pool;

[0049] Step 3: Receive configuration information, schedule a virtual cipher machine from the virtual cipher machine pool as a cipher machine node according to the configuration information, and build a virtual cipher machine cluster;

[0050] Step 4: Dynamically adjust the number of cipher machine nodes in the virtual cipher machine cluster according to the virtual cipher machine information and configuration information.

[0051] In a specific embodiment, a load balancing method applicable to a cloud server cryptographic machine, such as Figure 3 As shown, specifically including:

[0052] Step 1. When the cloud server cryptographic machine device is initialized, the device management module creates a specified number of virtual cryptographic machines and assigns independent IP addresses;

[0053] Step 2. The scheduling module detects the virtual cipher machine and cloud server cipher machine host in the local area network through UDP broadcast, and records the virtual cipher machine information consisting of the number of connections, service operation status, and cryptographic computing power utilization rate of the virtual cipher machine monitored by the device management module into the scheduling module as a virtual cipher machine pool, and sends the monitoring reporting address of the scheduling module to the above-mentioned device management module;

[0054] Step 3. The user directly operates the scheduling module through the interface, configures the cluster node threshold n, cryptographic computing power utilization coefficient M, service connection utilization coefficient N and other parameters of the scheduling module, and configures the cluster module load balancing mode; if the cluster module is deployed independently, configure the cluster module IP address in the scheduling module;

[0055] Step 4. When the cluster module is initialized, the scheduling module selects n+1 virtual cryptographic machines as nodes to build a cluster (n≥2), and the selected nodes span at least two cloud server cryptographic machine hosts;

[0056] Step 5. The device management module monitors the number of connections, service operation status, and cryptographic computing power usage of the virtual cryptographic machine and reports the information to the management module of the load balancing system;

[0057] Step 6. The scheduling module collects the virtual cipher machine information reported by the device management module and records it in the scheduling module;

[0058] Step 7. If a node fails, the scheduling module notifies the cluster module to remove the node and add a new node to the cluster; the scheduling module recycles the removed node to the virtual cryptographic machine pool, marks it as unavailable, and the node is no longer allocated until the device management module reports that the node is operating normally, marks it as available, and the node can be allocated again;

[0059] Step 8. The scheduling module determines the service connection utilization rate of each node N+ cryptographic computing power utilization Is M greater than or equal to 50%, where N+M=1? If so, expand the cluster and add a new node to the cluster; otherwise, determine whether the node service connection utilization rate is N+ cryptographic computing power utilization If M=0 and the number of nodes in the cluster is greater than n+1, the cluster will be scaled down and the node service connection usage will be removed. N+ cryptographic computing power utilization M=0; otherwise, the node is not adjusted; the scheduling module recycles the removed node to the virtual cipher machine pool, and the node can be reallocated.

[0060] Furthermore, it also includes step 9. The scheduling module compares the node pressure P of each node, dynamically modifies the weight of the nodes in the cluster, increases the weight of the node with a small node pressure P, and the server with a high weight value will process more requests, dynamically balancing the direct pressure of each node; for n+1 nodes, the node pressures are P1, P2, ..., Pn respectively. If , i=1,2,...,n+1, Pi represents the node pressure of the ith node, then the weight of the ith node is increased, otherwise the weight is reduced if it is greater than 0.

[0061] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0062] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A load balancing system suitable for cloud server cryptographic machines, It is characterized in that It includes a scheduling module and a cluster module; the scheduling module is connected to the cluster module; the scheduling module detects the cloud server cryptographic machine, obtains the virtual cryptographic machine information, and schedules the virtual cryptographic machine virtualized by the cloud server cryptographic machine to build a virtual cryptographic machine cluster according to the virtual cryptographic machine information and the received configuration information; The cluster module provides cryptographic operation services on behalf of the virtual cryptographic machine cluster, and the scheduling module notifies the cluster module to dynamically adjust the virtual cryptographic machine cluster according to the virtual cryptographic machine information and the configuration information; The scheduling module performs fault detection on the virtual cipher machine according to the virtual cipher machine information, and performs stress detection on the virtual cipher machine in combination with the configuration information, adds or removes the virtual cipher machine in the virtual cipher machine cluster according to the fault detection result and the stress detection result, and also adjusts the weight of the virtual cipher machine cluster according to the stress detection result; The cloud server cryptographic machine includes a device management module, which creates several virtual cryptographic machines and assigns a unique cryptographic machine IP to each virtual cryptographic machine, monitors the number of connections, service operation status and cryptographic computing power usage of all virtual cryptographic machines, and generates virtual cryptographic machine information to be transmitted to the scheduling module; The scheduling module detects the cloud server cryptographic machine in the local area network through UDP broadcast, records the virtual cryptographic machine information of all created virtual cryptographic machines to build a virtual cryptographic machine pool, and marks the call status of all the virtual cryptographic machines in the virtual cryptographic machine pool as available; When the cluster module is initialized, the scheduling module selects several virtual cryptographic machines from at least two cloud server cryptographic machines from the virtual cryptographic machine pool as cryptographic machine nodes to build a virtual cryptographic machine cluster; The number of the cryptographic machine nodes is a preset node threshold; The cluster module sends a heartbeat detection to all the cryptographic machine nodes to obtain a detection result. If the detection result is a fault type, the detection result is fed back to the scheduling module; The scheduling module traverses the virtual cipher machine information and generates the detection result when a faulty virtual cipher machine is found; The scheduling module notifies the cluster module to remove the corresponding cryptographic machine node from the virtual cryptographic machine pool and mark it as unavailable according to the detection result, and at the same time retrieves a corresponding number of available virtual cryptographic machines from the virtual cryptographic machine pool and adds them to the virtual cryptographic machine cluster; The scheduling module calculates the node pressure of each cryptographic machine node according to the virtual cryptographic machine information, and adds or removes the cryptographic machine node according to the comparison result between the node pressure and the set pressure threshold; the node pressure P expression is: P = service connection utilization rate N+ cryptographic computing power utilization M, wherein the service connection utilization rate is the ratio of the number of connections of the cryptographic machine node to the preset maximum number of connections, N represents the service connection utilization rate coefficient, M represents the cryptographic computing power utilization rate coefficient, and N+M=1; if the node pressure is greater than or equal to the pressure threshold, the scheduling module retrieves the newly available virtual cryptographic machine from the virtual cryptographic machine pool and adds it to the virtual cryptographic machine cluster; if the node pressure is 0 and the number of the cryptographic machine nodes is greater than the preset node threshold, the cryptographic machine node with the node pressure of 0 is removed from the virtual cryptographic machine pool; otherwise, the cryptographic machine node is not adjusted.

2. A load balancing system suitable for a cloud server cryptographic machine according to claim 1, It is characterized in that The cluster module records the weight and call status of each cryptographic machine node, acts as an agent for each cryptographic machine node, provides cryptographic operation services through a set service IP, and sends service requests to the corresponding cryptographic machine node for cryptographic operation according to the cryptographic machine IP.

3. A load balancing system suitable for a cloud server cryptographic machine according to claim 1, It is characterized in that It also includes an operation interface, which receives configuration information including node thresholds, service connection utilization coefficients and cryptographic computing power utilization coefficients, and the scheduling module dynamically adjusts the number of cryptographic machine nodes according to the configuration information.

4. A load balancing system suitable for a cloud server cryptographic machine according to claim 2, It is characterized in that The scheduling module calculates the node pressure of each cipher machine node according to the virtual cipher machine information. The node pressure P is expressed as: P = service connection utilization rate N+ cryptographic computing power utilization M, wherein the service connection utilization rate is the ratio of the number of connections of the cryptographic machine node to the preset maximum number of connections, N represents the service connection utilization rate coefficient, M represents the cryptographic computing power utilization rate coefficient, and N+M=1; the scheduling module compares the node pressure of each of the cryptographic machine nodes and increases the weight of the cryptographic machine nodes with small node pressure.

5. A load balancing method for cloud server cryptographic machines, It is characterized in that A load balancing system applicable to a cloud server cryptographic machine as described in any one of claims 1 to 4, comprising the following steps: Step 1: The cloud server cryptographic machine is initialized, and the device management module creates several virtual cryptographic machines; Step 2: Obtain the virtual cipher machine information of each virtual cipher machine and build a virtual cipher machine pool; Step 3: receiving configuration information, and scheduling the virtual cryptographic machine from the virtual cryptographic machine pool as a cryptographic machine node according to the configuration information to build a virtual cryptographic machine cluster; Step 4: Dynamically adjust the cryptographic machine nodes and quantity in the virtual cryptographic machine cluster according to the virtual cryptographic machine information and the configuration information.

Images