A cryptographic resource management and scheduling method and system

Abstract

The present application relates to the technical field of cryptographic resource management, and in particular to a cryptographic resource management scheduling method and system. It includes: establishing multiple service nodes; establishing virtual machines according to user password requirements, and generating allocation instructions according to all virtual machines to be deployed; establishing a prediction cycle, establishing an adjustment time node according to the prediction cycle, obtaining the real-time operating parameters of the service node according to the adjustment time node, and judging whether to generate a scheduling instruction. By dynamically adjusting the prediction cycle, the occupancy of each service node is predicted according to the adjustment time node, so as to formulate a scheduling plan in time to avoid excessive load on a single service node, resulting in reduced service quality. At the same time, by dynamically adjusting the prediction cycle, by dynamically adjusting the prediction cycle, and increasing the reserved space for each service node, the number of virtual machine migrations is reduced, thereby reducing the overall operating cost.

Description

Technical Field

[0001] The present application relates to the technical field of cryptographic resource management, and in particular to a cryptographic resource management scheduling method and system. Background Art

[0002] The power system is related to the national economy and people's livelihood. As the international cyber security situation becomes increasingly severe, the information security risks faced by the power grid are increasing, and there are security risks such as sensitive data leakage, illegal data tampering and identity fraud. Passwords are the basis for building a power network security protection system.

[0003] In order to solve the problem of massive data encryption and user identity authentication in cloud computing environments, traditional discrete cryptographic servers can no longer meet the needs. It is necessary to integrate cryptographic servers with specialized hardware security modules that meet the requirements to form a cryptographic resource pool, which will be centrally scheduled and managed by the cloud platform to meet the needs of large-scale cryptographic operations. In order to solve the problem of cryptographic security and efficient operation of massive concurrent services, it is urgent to study the collaborative dynamic scheduling technology of cryptographic resources to realize the dynamic allocation and on-demand adjustment of cryptographic resources according to business traffic. Summary of the invention

[0004] The purpose of this application is to solve the above technical problems and provide a cryptographic resource management and scheduling method and system, aiming to achieve efficient allocation of cryptographic resources and ensure load balancing of each resource node.

[0005] In some embodiments of the present application, the occupancy of each service node is predicted according to the adjustment time node by dynamically adjusting the prediction cycle, so as to formulate a scheduling plan in time to avoid excessive load on a single service node, which causes a decrease in service quality. At the same time, by dynamically adjusting the prediction cycle and increasing the reserved space for each service node, the number of virtual machine migrations is reduced, thereby reducing the overall operating cost.

[0006] In some embodiments of the present application, multiple allocation plans are generated based on the parameters of the virtual machines to be deployed, and a secondary optimization process is performed, thereby improving the efficiency of resource utilization and reducing overall operating energy consumption on the basis of balancing the load of each service node, while avoiding downtime due to sudden loads and improving service quality.

[0007] In some embodiments of the present application, a cryptographic resource management scheduling method is provided, including:

[0008] Establish multiple service nodes;

[0009] Create virtual machines based on user password requirements and generate allocation instructions based on all virtual machines to be deployed;

[0010] A prediction cycle is established, an adjustment time node is established according to the prediction cycle, real-time operation parameters of the service node are obtained according to the adjustment time node, and it is determined whether to generate a scheduling instruction.

[0011] In some embodiments of the present application, when generating allocation instructions according to all virtual machines to be deployed, it includes:

[0012] Get all virtual machines to be deployed;

[0013] Generate the resource requirement sequence A=(a1,a2…a n ), where ai is the resource requirement of the i-th virtual machine to be deployed, and n is the number of virtual machines to be deployed;

[0014] Establishing an allocation model, and generating a plurality of primary allocation plans according to the resource demand quantity series A and the allocation model;

[0015] Set the remaining resource threshold of each service node;

[0016] Screening all the primary allocation plans according to the remaining resource threshold to generate multiple secondary allocation plans;

[0017] Generate operational evaluation values ​​for each secondary allocation plan;

[0018] Establish the running evaluation value series C, C=(c1,c2…c n1 ), where n1 is the number of secondary allocation plans, and ci is the operation evaluation value of the i-th secondary allocation plan;

[0019] Get the maximum value c in the running evaluation value sequence C max ;

[0020] According to the c max The corresponding secondary allocation plan generates allocation instructions.

[0021] In some embodiments of the present application, the generating of the operation evaluation value of each secondary allocation plan includes:

[0022] Obtain the occupied number m1 of service nodes in a single secondary allocation plan, the load parameters and resource occupancy of each service node;

[0023] Generating the operating cost evaluation value D1 of the secondary allocation plan according to the occupied quantity m1;

[0024] Generate a load evaluation value D2 of the secondary allocation plan based on the load parameters of each service node;

[0025] Generate resource utilization evaluation value D3 of the secondary allocation plan according to the resource occupancy of each service node;

[0026] Generate the operational evaluation value c of the secondary allocation plan;

[0027] c=e1*D1+e2*D2+e3*D3, wherein e1 is the first weight coefficient, e2 is the preset second weight coefficient, e3 is the preset third weight coefficient, and e1+e2+e3=1.

[0028] In some embodiments of the present application, the establishment of the prediction cycle includes:

[0029] Get the number of virtual machines on the service node and the resource usage of each virtual machine;

[0030] Generate a virtual machine evaluation value f of the service node;

[0031] Preset a first virtual machine evaluation value interval (F1, F2), a second virtual machine evaluation value interval (F2, F3) and a third virtual machine evaluation value interval (F3, F4);

[0032] The first prediction cycle duration T1, the second prediction cycle duration T2 and the third prediction cycle duration T3 are preset, and T1 <T2<T3;

[0033] Set the prediction period t according to the number of virtual machines f;

[0034] If the number of virtual machines f is in the first virtual machine evaluation value interval, the prediction period t is set to the preset first prediction period T1, that is, t=T1;

[0035] If the number of virtual machines f is in the second virtual machine evaluation value interval, the prediction period t is set to the preset second prediction period T2, ie t = T2;

[0036] If the number of virtual machines f is in the third virtual machine evaluation value interval, the prediction cycle duration t is set to the preset third prediction cycle duration T1, that is, t=T3.

[0037] In some embodiments of the present application, the determining whether to generate a scheduling instruction includes:

[0038] Establish a cryptographic resource demand prediction model;

[0039] Generate the expected total demand of all virtual machines of the service node in the next forecast period according to the cryptographic resource demand forecast model

[0040] Acquire the resource amount of the service node, and generate the remaining resource amount b of the service node in the next prediction period according to the expected total demand and the resource amount;

[0041] A first remaining resource amount B1 and a second remaining resource amount B2 are preset, and B1 <B2;

[0042] If b < B1, set the service node as a first-level service node;

[0043] If B1 < b < B2, set the service node as a second-level service node;

[0044] If b > B2, set the service node as a third-level service node;

[0045] If the current service node is a first-level service node, generate a scheduling instruction.

[0046] In some embodiments of the present application, when generating the scheduling instruction, it includes:

[0047] Obtain all virtual machine parameters of a single first-level service node, generate the occupied resource amount, remaining running time, and scheduling times of each virtual machine;

[0048] Generate an eviction evaluation value of the virtual machine according to the remaining running time and the scheduling times;

[0049] Generate an eviction order of the virtual machine according to the eviction evaluation value;

[0050] Generate the resource amount b1 to be evicted from the first-level service node, and select the virtual machine to be evicted according to the virtual machine eviction order and the resource amount b1 to be evicted;

[0051] Obtain all virtual machine parameters to be evicted, and generate a scheduling plan according to all the virtual machine parameters to be evicted.

[0052] In some embodiments of the present application, a password resource management and scheduling system is provided, including:

[0053] A central control unit for establishing multiple service nodes;

[0054] A demand unit connected to the central control unit through a wireless signal. The demand unit is used to establish a virtual machine according to the user's password demand and send the virtual machine parameters to the central control unit. The central control unit generates an allocation instruction according to all the virtual machines to be deployed;

[0055] A scheduling unit for establishing a prediction period. The scheduling unit establishes an adjustment time node according to the prediction period, obtains the real-time operation parameters of the service node according to the adjustment time node, and determines whether to generate a scheduling instruction.

[0056] In some embodiments of the present application, the central control unit includes:

[0057] A receiving module for obtaining all the virtual machines to be deployed and generating a sequence A=(a1, a2…a n ) of the resource demand amounts of the virtual machines to be deployed, where ai is the resource demand amount of the i-th virtual machine to be deployed;

[0058] A first processing module is used to establish an allocation model and generate a plurality of primary allocation plans according to the resource demand quantity series A and the allocation model;

[0059] A second processing module is used to set a remaining resource threshold of each service node, screen all the primary allocation plans according to the remaining resource threshold, and generate multiple secondary allocation plans;

[0060] The third processing module is used to generate the operation evaluation value of each secondary allocation plan and establish the operation evaluation value sequence C, C = (c1, c2…c n1 ), where n1 is the number of secondary allocation plans, and ci is the operation evaluation value of the i-th secondary allocation plan;

[0061] The third processing module is also used to obtain the maximum value c in the running evaluation value sequence C ma x, according to the c max The corresponding secondary allocation plan generates allocation instructions.

[0062] In some embodiments of the present application, the scheduling unit includes:

[0063] A first scheduling module is used to obtain the number of virtual machines of the service node and the resource usage of each virtual machine, and generate a virtual machine evaluation value f of the service node. The first scheduling module also sets a prediction cycle duration t according to the number of virtual machines f in English;

[0064] If the number of virtual machines f is in the first virtual machine evaluation value interval, the prediction period t is set to the preset first prediction period T1, that is, t=T1;

[0065] If the number of virtual machines f is in the second virtual machine evaluation value interval, the prediction period t is set to the preset second prediction period T2, ie t = T2;

[0066] If the number of virtual machines f is in the third virtual machine evaluation value interval, the prediction cycle duration t is set to the preset third prediction cycle duration T1, that is, t=T3.

[0067] In some embodiments of the present application, the scheduling unit further includes:

[0068] The prediction module is used to establish a cryptographic resource demand prediction model, and generate the expected total demand of all virtual machines of the service node in the next prediction period according to the cryptographic resource demand prediction model.

[0069] A second scheduling module is used to obtain the resource amount of the service node, and generate the remaining resource amount b of the service node in the next prediction period according to the expected total demand and the resource amount;

[0070] The second scheduling module is further used to set the type of the current service node according to the remaining resource amount b;

[0071] The third scheduling module generates a scheduling instruction if the current service node is a first-level service node.

[0072] Compared with the prior art, the cryptographic resource management scheduling method and system of the embodiment of the present application has the following beneficial effects:

[0073] By dynamically adjusting the prediction cycle, the occupancy of each service node is predicted according to the adjustment time node, so as to formulate a scheduling plan in time to avoid excessive load on a single service node, which will cause reduced service quality. At the same time, by dynamically adjusting the prediction cycle, the reserved space is increased for each service node, the number of virtual machine migrations is reduced, thereby reducing the overall operating cost.

[0074] Generate multiple allocation plans based on the parameters of the virtual machines to be deployed, and perform secondary optimization processing, so as to improve the efficiency of resource utilization and reduce the overall operating energy consumption on the basis of balancing the load of each service node, while avoiding downtime due to sudden loads and improving service quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a flow chart of a cryptographic resource management scheduling method in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0076] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0077] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0078] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0079] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0080] like Figure 1 As shown, a cryptographic resource management scheduling method of a preferred embodiment of the present application includes:

[0081] S101: Establish multiple service nodes;

[0082] S102: Establish a virtual machine according to the user password requirement, and generate allocation instructions according to all virtual machines to be deployed;

[0083] S103: Establish a prediction cycle, establish an adjustment time node according to the prediction cycle, obtain the real-time operation parameters of the service node according to the adjustment time node, and determine whether to generate a scheduling instruction.

[0084] Specifically, when generating allocation instructions based on all virtual machines to be deployed, it includes:

[0085] Get all virtual machines to be deployed;

[0086] Generate the resource requirement sequence A=(a1,a2…a n ), where ai is the resource requirement of the i-th virtual machine to be deployed, and n is the number of virtual machines to be deployed;

[0087] Establish an allocation model, and generate multiple primary allocation plans according to the resource demand series A and the allocation model;

[0088] Set the remaining resource threshold of each service node;

[0089] Screen all primary allocation plans according to the remaining resource threshold and generate multiple secondary allocation plans;

[0090] Generate operational evaluation values ​​for each secondary allocation plan;

[0091] Establish the running evaluation value series C, C=(c1,c2…c n1 ), where n1 is the number of secondary allocation plans, and ci is the operation evaluation value of the i-th secondary allocation plan;

[0092] Get the maximum value c in the running evaluation value sequence C max ;

[0093] According to c max The corresponding secondary allocation plan generates allocation instructions.

[0094] Specifically, the remaining resource threshold of each service node can be set according to historical operation data. The resource space reserved for each service node is mainly used to deal with sudden load problems and avoid downtime, thereby ensuring service quality.

[0095] Specifically, when all primary allocation plans are screened according to the remaining resource threshold, it means judging whether there is a service node with remaining resources less than the remaining resource threshold according to the primary allocation plan value. If so, the primary allocation plan is eliminated; if not, a secondary allocation plan is generated according to the current initial allocation plan.

[0096] Specifically, when generating the operational evaluation value of each secondary allocation plan, it includes:

[0097] Obtain the occupied number m1 of service nodes in a single secondary allocation plan, the load parameters and resource occupancy of each service node;

[0098] Generate the operating cost evaluation value D1 of the secondary allocation plan according to the occupied quantity m1;

[0099] Generate a load evaluation value D2 of the secondary allocation plan based on the load parameters of each service node;

[0100] Generate resource utilization evaluation value D3 of the secondary allocation plan according to the resource occupancy of each service node;

[0101] Generate the operational evaluation value c of the secondary allocation plan;

[0102] c=e1*D1+e2*D2+e3*D3, wherein e1 is the first weight coefficient, e2 is the preset second weight coefficient, e3 is the preset third weight coefficient, and e1+e2+e3=1.

[0103] Specifically, the first weight coefficient, the second weight coefficient and the third weight coefficient can be set according to the optimization direction. The optimization direction includes the load balancing goal, and the second weight coefficient is larger under this goal. The optimization direction also includes the goal of improving resource utilization, and the third weight coefficient is larger under this goal. The optimization direction also includes the goal of reducing energy consumption, and the first weight coefficient is larger under this goal.

[0104] Specifically, the smaller the average load, the higher the corresponding load evaluation value. At the same time, the load evaluation value is corrected according to the load difference of each node. The smaller the load difference of each node, the higher the corresponding load evaluation value.

[0105] Specifically, the fewer service nodes are occupied, the higher the operating cost evaluation value is; the greater the resource occupation of the service nodes is, the higher the corresponding resource utilization efficiency evaluation value is.

[0106] Specifically, the operating cost evaluation value, the load evaluation value and the resource utilization efficiency evaluation value have the same value range.

[0107] It can be understood that in the above embodiment, multiple allocation plans are generated according to the parameters of the virtual machines to be deployed, and a secondary optimization process is performed, so as to improve the efficiency of resource utilization and reduce the overall operating energy consumption on the basis of balancing the load of each service node, while avoiding downtime due to sudden loads and improving service quality.

[0108] In a preferred embodiment of the present application, when establishing a prediction cycle, it includes:

[0109] Get the number of virtual machines on the service node and the resource usage of each virtual machine;

[0110] Generate a virtual machine evaluation value f of the service node;

[0111] Preset a first virtual machine evaluation value interval (F1, F2), a second virtual machine evaluation value interval (F2, F3) and a third virtual machine evaluation value interval (F3, F4);

[0112] The first prediction cycle duration T1, the second prediction cycle duration T2 and the third prediction cycle duration T3 are preset, and T1 <T2<T3;

[0113] Set the prediction period t according to the number of virtual machines f;

[0114] If the number of virtual machines f is in the first virtual machine evaluation value interval, the prediction period t is set to the preset first prediction period T1, that is, t=T1;

[0115] If the number of virtual machines f is in the second virtual machine evaluation value interval, the prediction period t is set to the preset second prediction period T2, ie t = T2;

[0116] If the number of virtual machines f is in the third virtual machine evaluation value interval, the prediction cycle duration t is set to the preset third prediction cycle duration T1, that is, t=T3.

[0117] Specifically, the total resource occupancy of the service node at the current adjustment time node is generated. The larger the total resource occupancy, the smaller the corresponding initial virtual machine evaluation value. At the same time, the initial virtual machine evaluation value is corrected according to the number of virtual machines at the current service node. The more virtual machines there are, the higher the corrected virtual machine evaluation value.

[0118] It is understandable that in the above embodiments, the prediction cycle duration is dynamically adjusted by generating virtual machine evaluation values, thereby ensuring the operational stability of the service node within a single prediction cycle and improving the service quality.

[0119] In a preferred embodiment of the present application, when determining whether to generate a scheduling instruction, it includes:

[0120] Establish a password resource demand prediction model;

[0121] Generate the total expected demand for all virtual machines of the service node in the next prediction cycle according to the password resource demand prediction model

[0122] Obtain the resource amount of the service node, and generate the remaining resource amount b of the service node in the next prediction cycle according to the total expected demand and the resource amount;

[0123] Preset a first remaining resource amount B1 and a second remaining resource amount B2, and B1 < B2;

[0124] If b < B1, set the service node as a first-level service node;

[0125] If B1 < b < B2, set the service node as a second-level service node;

[0126] If b > B2, set the service node as a third-level service node;

[0127] If the current service node is a first-level service node, generate a scheduling instruction.

[0128] Specifically, when generating a scheduling instruction, it includes:

[0129] Obtain all virtual machine parameters of a single first-level service node, and generate the occupied resource amount, remaining running time, and scheduling times of each virtual machine;

[0130] Generate an eviction evaluation value of the virtual machine according to the remaining running time and the scheduling times;

[0131] Generate the eviction order of the virtual machines according to the eviction evaluation value;

[0132] Generate the resource amount b1 to be evicted from the first-level service node, and select the virtual machines to be evicted according to the eviction order of the virtual machines and the resource amount b1 to be evicted;

[0133] Obtain all virtual machine parameters to be evicted, and generate a scheduling plan according to all virtual machine parameters to be evicted.

[0134] Specifically, the first-level service node means that the resource amount of the current service node is insufficient, the second-level service node means that the current service node is operating normally, and the third-level service node means that the resource of the current service node is excessive.

[0135] Specifically, the shorter the remaining running time is, the higher the call-out evaluation value is. At the same time, the call-out evaluation value is corrected according to the number of calls. The more the number of calls is, the smaller the corrected running evaluation value is.

[0136] Specifically, the remaining running time refers to the time required for the current virtual machine to finish running, and the number of calls refers to the number of times the current virtual machine changes the service node.

[0137] Specifically, after all the virtual machines that need to be called out are selected, the remaining resources of all service nodes are updated, and multiple primary scheduling plans are generated based on the updated remaining resources and all the virtual machines that need to be called out. After optimization processing, the scheduling plan is determined. If the current service node cannot meet the resource requirements of all the virtual machines that need to be called out, the non-running service node can be enabled.

[0138] It can be understood that in the above embodiment, multiple allocation plans are generated according to the parameters of the virtual machines to be deployed, and a secondary optimization process is performed, so as to improve the efficiency of resource utilization and reduce the overall operating energy consumption on the basis of balancing the load of each service node, while avoiding downtime due to sudden loads and improving service quality.

[0139] Based on another preferred embodiment of a cryptographic resource management scheduling method in any of the above preferred embodiments, this preferred embodiment provides a cryptographic resource management scheduling system, including:

[0140] Central control unit, used to establish multiple service nodes;

[0141] The demand unit is connected to the central control unit via wireless signals. The demand unit is used to establish a virtual machine according to the user password requirements and send the virtual machine parameters to the central control unit. The central control unit generates allocation instructions according to all virtual machines to be deployed.

[0142] The scheduling unit is used to establish a prediction cycle. The scheduling unit establishes an adjustment time node according to the prediction cycle, obtains the real-time operating parameters of the service node according to the adjustment time node, and determines whether to generate a scheduling instruction.

[0143] Specifically, the central control unit includes:

[0144] The receiving module is used to obtain all virtual machines to be deployed and generate a resource demand sequence A=(a1, a2…a n ), where ai is the resource requirement of the i-th virtual machine to be deployed;

[0145] The first processing module is used to establish an allocation model and generate multiple primary allocation plans according to the resource demand quantity series A and the allocation model;

[0146] The second processing module is used to set the remaining resource threshold of each service node, screen all primary allocation plans according to the remaining resource threshold, and generate multiple secondary allocation plans;

[0147] The third processing module is used to generate the operation evaluation value of each secondary allocation plan and establish the operation evaluation value sequence C, C = (c1, c2…c n1 ), where n1 is the number of secondary allocation plans, and ci is the operation evaluation value of the i-th secondary allocation plan;

[0148] The third processing module is also used to obtain the maximum value c in the running evaluation value sequence C max , according to c max The corresponding secondary allocation plan generates allocation instructions.

[0149] Specifically, the scheduling unit includes:

[0150] A first scheduling module is used to obtain the number of virtual machines of the service node and the resource usage of each virtual machine, and generate a virtual machine evaluation value f of the service node. The first scheduling module also sets a prediction cycle duration t according to the number of virtual machines f in English;

[0151] If the number of virtual machines f is in the first virtual machine evaluation value interval, the prediction period t is set to the preset first prediction period T1, that is, t=T1;

[0152] If the number of virtual machines f is in the second virtual machine evaluation value interval, the prediction period t is set to the preset second prediction period T2, ie t = T2;

[0153] If the number of virtual machines f is in the third virtual machine evaluation value interval, the prediction cycle duration t is set to the preset third prediction cycle duration T1, that is, t=T3.

[0154] The prediction module is used to establish a cryptographic resource demand prediction model and generate the expected total demand for all virtual machines of the service node in the next prediction period according to the cryptographic resource demand prediction model.

[0155] The second scheduling module is used to obtain the resource amount of the service node and generate the remaining resource amount b of the service node in the next prediction period according to the expected total demand and resource amount;

[0156] The second scheduling module is also used to set the type of the current service node according to the remaining resource amount b;

[0157] The third scheduling module generates a scheduling instruction if the current service node is a first-level service node.

[0158] According to the first concept of the present application, the occupancy of each service node is predicted according to the adjustment time node by dynamically adjusting the prediction cycle, so as to formulate a scheduling plan in time to avoid excessive load on a single service node, which causes a decrease in service quality. At the same time, by dynamically adjusting the prediction cycle and increasing the reserved space for each service node, the number of virtual machine migrations can be reduced, thereby reducing the overall operating cost.

[0159] According to the second concept of the present application, multiple allocation plans are generated according to the parameters of the virtual machines to be deployed, and secondary optimization processing is performed, so as to improve the efficiency of resource utilization and reduce the overall operating energy consumption on the basis of balancing the load of each service node, while avoiding downtime due to sudden loads and improving service quality.

[0160] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present application. These improvements and substitutions should also be regarded as the scope of protection of the present application.

Claims

1. A cryptographic resource management scheduling method, characterized in that: Including: Establish multiple service nodes; Create virtual machines according to the user's password requirements, and generate allocation instructions based on all virtual machines to be deployed; Establish a prediction period, establish adjustment time nodes according to the prediction period, obtain the real-time operation parameters of service nodes according to the adjustment time nodes, and determine whether to generate scheduling instructions; When generating allocation instructions based on all virtual machines to be deployed, it includes: Obtain all virtual machines to be deployed; Generate the resource requirement sequence A=(a1,a2…a n ), where ai is the resource requirement of the i-th virtual machine to be deployed, and n is the number of virtual machines to be deployed; Establish an allocation model, and generate multiple primary allocation plans according to the resource demand sequence A and the allocation model; Set the remaining resource quantity thresholds for each service node; Screen all the primary allocation plans according to the remaining resource quantity thresholds to generate multiple secondary allocation plans; Generate the operation evaluation values for each secondary allocation plan; Establish the running evaluation value series C, C=(c1,c2…c n1 ), where n1 is the number of secondary allocation plans, and ci is the operation evaluation value of the i-th secondary allocation plan; Get the maximum value c in the running evaluation value sequence C max ; According to the c max The corresponding secondary allocation plan generates allocation instructions; When generating the operation evaluation values for each secondary allocation plan, it includes: Obtain the occupied quantity m1 of service nodes in a single secondary allocation plan, the load parameters and resource occupancy of each service node; Generate the operation cost evaluation value D1 of the secondary allocation plan according to the occupied quantity m1; Generate the load evaluation value D2 of the secondary allocation plan according to the load parameters of each service node; Generate the resource utilization evaluation value D3 of the secondary allocation plan according to the resource occupancy of each service node; Generate the operation evaluation value c of the secondary allocation plan; c = e1*D1 + e2*D2 + e3*D3, where e1 is the first weight coefficient, e2 is the preset second weight coefficient, e3 is the preset third weight coefficient, and e1 + e2 + e3 = 1.

2. The cryptographic resource management scheduling method according to claim 1, characterized in that: When establishing the prediction period, it includes: Obtain the number of virtual machines of the service node and the resource occupancy of each virtual machine; Generate the virtual machine evaluation value f of the service node; Preset the first virtual machine evaluation value interval (F1, F2), the second virtual machine evaluation value interval (F2, F3), and the third virtual machine evaluation value interval (F3, F4); Preset the first prediction period duration T1, the second prediction period duration T2, and the third prediction period duration T3, and T1 < T2 < T3; Set the prediction period duration t according to the number of virtual machines f; If the number of virtual machines f is within the first virtual machine evaluation value interval, set the prediction period duration t as the preset first prediction period duration T1, that is, t = T1; If the number of virtual machines f is within the second virtual machine evaluation value interval, set the prediction period duration t as the preset second prediction period duration T2, that is, t = T2; If the number of virtual machines f is within the third virtual machine evaluation value interval, set the prediction period duration t as the preset third prediction period duration T1, that is, t = T3.

3. The cryptographic resource management scheduling method according to claim 2, characterized in that: When determining whether to generate scheduling instructions, it includes: Establish a password resource demand prediction model; Generate the total expected demand of all virtual machines of the service node in the next prediction period according to the password resource demand prediction model Obtain the resource quantity of the service node, and generate the remaining resource quantity b of the service node in the next prediction period according to the total expected demand and the resource quantity; Preset the first remaining resource quantity B1 and the second remaining resource quantity B2, and B1 < B2; If b < B1, set the service node as a first-level service node; If B1 < b < B2, set the service node as a second-level service node; If b>B2, set the service node to a level 3 service node; If the current service node is a first-level service node, a scheduling instruction is generated.

4. The cryptographic resource management scheduling method according to claim 3, characterized in that: The generation of the scheduling instruction includes: Obtain all virtual machine parameters of a single first-level service node, generate the resource usage, remaining running time and scheduling times of each virtual machine; Generate a call-out evaluation value of the virtual machine according to the remaining running time and the scheduling number; Generate a virtual machine call-out sequence according to the call-out evaluation value; Generate the amount of resources b1 that need to be transferred out of the first-level service node, and select the virtual machine to be transferred out according to the virtual machine transfer order and the amount of resources b1 that need to be transferred out; All the parameters of the virtual machines that need to be called out are obtained, and a scheduling plan is generated according to all the parameters of the virtual machines that need to be called out.

5. A cryptographic resource management and scheduling system, using the cryptographic resource management and scheduling method according to any one of claims 1 to 4, characterized in that: include: Central control unit, used to establish multiple service nodes; A demand unit connected to the central control unit via a wireless signal, the demand unit is used to establish a virtual machine according to the user password requirement and send the virtual machine parameters to the central control unit, and the central control unit generates an allocation instruction according to all the virtual machines to be deployed; A scheduling unit, configured to establish a prediction cycle, wherein the scheduling unit establishes an adjustment time node according to the prediction cycle, obtains real-time operating parameters of the service node according to the adjustment time node, and determines whether to generate a scheduling instruction; The central control unit comprises: The receiving module is used to obtain all virtual machines to be deployed and generate a resource demand sequence A=(a1, a2…a n ), where ai is the resource requirement of the i-th virtual machine to be deployed; A first processing module is used to establish an allocation model and generate a plurality of primary allocation plans according to the resource demand quantity series A and the allocation model; A second processing module is used to set a remaining resource threshold of each service node, screen all the primary allocation plans according to the remaining resource threshold, and generate multiple secondary allocation plans; The third processing module is used to generate the operation evaluation value of each secondary allocation plan and establish the operation evaluation value sequence C, C = (c1, c2…c n1 ), where n1 is the number of secondary allocation plans, and ci is the operation evaluation value of the i-th secondary allocation plan; The third processing module is also used to obtain the maximum value c in the running evaluation value sequence C max , according to the cmax The corresponding secondary allocation plan generates allocation instructions.

6. The cryptographic resource management and scheduling system according to claim 5, characterized in that: The scheduling unit comprises: A first scheduling module is used to obtain the number of virtual machines of the service node and the resource usage of each virtual machine, and generate a virtual machine evaluation value f of the service node. The first scheduling module also sets a prediction cycle duration t according to the number of virtual machines f in English; If the number of virtual machines f is in the first virtual machine evaluation value interval, the prediction period t is set to the preset first prediction period T1, that is, t=T1; If the number of virtual machines f is in the second virtual machine evaluation value interval, the prediction period t is set to the preset second prediction period T2, ie t = T2; If the number of virtual machines f is in the third virtual machine evaluation value interval, the prediction cycle duration t is set to the preset third prediction cycle duration T1, that is, t=T3.

7. The cryptographic resource management and scheduling system according to claim 6, characterized in that: The scheduling unit also includes: The prediction module is used to establish a cryptographic resource demand prediction model, and generate the expected total demand of all virtual machines of the service node in the next prediction period according to the cryptographic resource demand prediction model. A second scheduling module is used to obtain the resource amount of the service node, and generate the remaining resource amount b of the service node in the next prediction period according to the expected total demand and the resource amount; The second scheduling module is further used to set the type of the current service node according to the remaining resource amount b; The third scheduling module generates a scheduling instruction if the current service node is a first-level service node.

Images