Data center infrastructure personnel intelligent configuration system

Abstract

The invention discloses a data center infrastructure personnel intelligent configuration system, and belongs to the technical field of data center operation and maintenance management. Comprising an acquisition module, a man-hour prediction module, a configuration module and a feedback module, the man-hour prediction module is used for calculating the health coefficient of the equipment according to the equipment list and performing man-hour prediction; the configuration module is used for configuring the number of personnel according to the prediction result; the feedback module is used for outputting an executable personnel configuration scheme through a preset scheduling system; according to the method, a multi-dimensional configuration scheme meeting the SLA requirement is output through a real-time optimization algorithm of precise man-hour modeling and fusion of dynamic factors, real-time generation of the scheme is achieved, and traditional manual measurement and calculation which consumes several days are replaced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data center operation and maintenance management, in particular to a data center infrastructure personnel intelligent configuration system. BACKGROUND

[0002] Currently, personnel configuration of data centers mainly relies on two types of static experience methods: one is a linear formula method based on the number of devices, such as one person for every N cold machines, and the other is an estimation method based on cabinet density, such as one person managing M cabinets. These methods have significant defects: first, they are static models that cannot reflect dynamic factors such as device aging, failure rate changes, and automation level improvements, resulting in a large deviation between the calculated results and actual needs; second, they lack precise quantification of device-level operation and maintenance hours, and can only make coarse-grained estimates; third, they cannot flexibly adapt to the differentiated needs of different customers in terms of service level, cost control, and technical foresight.

[0003] Therefore, there is an urgent need in the field for an intelligent personnel configuration system that can dynamically adapt, accurately quantify, and output diversified solutions. SUMMARY

[0004] The purpose of the present application is to provide a data center infrastructure personnel intelligent configuration system that can generate optimal and diversified personnel configuration schemes in real time based on the actual state of the devices, automation level, and specific SLA requirements, to solve the problems raised in the background art.

[0005] To solve the above technical problems, the present application provides the following technical solutions: The data center infrastructure personnel intelligent configuration system includes a collection module, an operation hour prediction module, a configuration module, and a feedback module: The collection module is used to obtain a device list and extract dynamic factors; The operation hour prediction module is used to predict operation hours based on the device list; The configuration module is used to configure the number of personnel based on the prediction results; The feedback module is used to output an executable personnel configuration scheme through a pre-set scheduling system.

[0006] As a preferred solution of the data center infrastructure personnel intelligent configuration system of the present application, the collection module includes real-time acquisition of a device list of the data center infrastructure when a system start instruction is triggered, and the device list record includes structured data of device type, device model, device quantity, device installation location, commissioning date, rated parameters, maintenance status, and historical fault codes; dynamic factors are extracted according to the type of the data center, including SLA weight and operation and maintenance mode.

[0007] As a preferred scheme of the data center infrastructure personnel intelligent configuration system, the man-hour prediction module comprises a preprocessing unit, a similarity unit and a prediction unit. The preprocessing unit is configured to extract corresponding equipment standard service life and basic man-hour from a preset knowledge base according to the equipment model, and calculate a health coefficient corresponding to the equipment model. The similarity unit is configured to extract the equipment standard service life and basic man-hour corresponding to the equipment model of the equipment of the similar equipment type when the equipment model does not exist in the preset knowledge base. The prediction unit is configured to construct a man-hour prediction model, input the equipment model, the equipment quantity, the basic man-hour and the health coefficient, and output total man-hour, and the prediction formula is:

[0008] wherein, total man-hour is represented by T, the number of equipment models is represented by N, the equipment quantity of the i-th equipment model is represented by Ni, the basic man-hour of the i-th equipment model is represented by Bi, and the health coefficient of the i-th equipment model is represented by Hi.

[0009] As a preferred scheme of the data center infrastructure personnel intelligent configuration system, the specific implementation process of calculating the health coefficient corresponding to the equipment model comprises: According to the commissioning date of the equipment model, the actual service life of the equipment model is obtained, and the life coefficient of the equipment model is calculated , and the calculation formula is:

[0010] wherein, Y represents the actual service life of the equipment model, the health coefficient increment of the equipment model is represented by ΔHi; the health coefficient increment is determined by a preset health coefficient increment template, and the health coefficient increment template comprises a mapping rule of the ratio of the actual service life of the equipment model to the standard service life and the health coefficient increment; The number of failures of the equipment model within a time period is counted, and a failure coefficient of the equipment model is calculated , and the calculation formula is:

[0011] wherein, R represents the number of failures of the equipment model within a time period, and the preset coefficient is represented by K; Based on the operating environment of the equipment, the environment coefficient of the equipment model is extracted from a preset environment coefficient table. The health score of the equipment model is obtained according to the product of the age coefficient, the failure coefficient and the environment coefficient of the equipment model; The maintenance coefficient, the location coefficient and the cluster coefficient are extracted from a preset dynamic correction table according to the maintenance state, the equipment installation position and the equipment quantity of the equipment model; The product of the health score of the equipment model, the maintenance coefficient, the location coefficient and the cluster coefficient is taken as the health coefficient corresponding to the equipment model.

[0012] As a preferred scheme of the intelligent personnel configuration system of the data center infrastructure, the selection process of the equipment of the similar equipment type comprises: The equipment type, the rated parameter, the manufacturer, the complexity coefficient and the key technology of the equipment are obtained, the similarity of the equipment and the equipment of different equipment models in the preset knowledge base is calculated, and the calculation formula is:

[0013]

[0014] wherein, the similarity of the equipment and the equipment of the jth equipment model in the preset knowledge base, a value for reflecting whether the data types of the two equipment are same, a value for reflecting the closeness of the rated parameters of the two equipment, the reciprocal of the number of different key technologies of the equipment and the equipment of the jth equipment model in the preset knowledge base, and when the number of different key technologies is 0, the value is 1, a value for reflecting whether the manufacturers of the two equipment are same, the complexity coefficient, Q represents the number of rated parameters, the parameter value of the xth rated parameter of the equipment, the parameter value of the xth rated parameter of the equipment of the jth equipment model, a weight; The equipment models are sorted according to the similarity from large to small to generate a similar sequence, and the equipment of the first equipment model is selected as the equipment of the similar equipment type.

[0015] As a preferred scheme of the intelligent personnel configuration system of the data center infrastructure, the configuration module comprises: The automation replacement rate ple is extracted according to the operation and maintenance mode, the personnel working hour basis SP of the data center is obtained, and the final number of configured personnel GH is calculated, and the calculation formula is:

[0016] wherein, represents the SLA weight.

[0017] As a preferred solution of the data center infrastructure personnel intelligent configuration system of the present application, the SLA weight is a redundancy amplification coefficient of a service level agreement requirement of the data center.

[0018] As a preferred solution of the data center infrastructure personnel intelligent configuration system of the present application, the preset scheduling system includes an input final configuration personnel number and a trigger output executable personnel configuration scheme generation instruction.

[0019] Compared with the prior art, the beneficial effects achieved by the present application are: the data center infrastructure personnel intelligent configuration system provided by the present application includes a collection module, a work hour prediction module, a configuration module and a feedback module: the collection module is used to obtain a device list and extract dynamic factors; the work hour prediction module is used to perform work hour prediction according to the device list; the configuration module is used to configure the number of personnel according to the prediction result; and the feedback module is used to output an executable personnel configuration scheme through a preset scheduling system; the present application outputs a multi-dimensional configuration scheme meeting the SLA requirement through accurate work hour modeling and real-time optimization algorithm of dynamic factors, realizes real-time generation of the scheme, and replaces the traditional time-consuming manual calculation of several days. BRIEF DESCRIPTION OF DRAWINGS

[0020] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, illustrate the present application, and are used together with the embodiments of the present application to explain the present application, and do not constitute a limitation on the present application.

[0021] Figure 1 is a structural schematic diagram of the data center infrastructure personnel intelligent configuration system of the present application; Figure 2 is structured data in the embodiment of the present application; Figure 3 is a health coefficient increment template in the embodiment of the present application; Figure 4 is a matching instruction in the embodiment of the present application; Figure 5 is an automation replacement rate in the embodiment of the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0023] Please refer toFigure 1 In the first embodiment, an intelligent configuration system for data center infrastructure personnel is provided, characterized in that the system comprises a collection module, a work hour prediction module, a configuration module, and a feedback module: The collection module is configured to obtain a device list and extract dynamic factors. The work hour prediction module is configured to predict work hours based on the device list. The configuration module is configured to configure the number of personnel based on the prediction results. The feedback module is configured to output an executable personnel configuration scheme through a preset scheduling system.

[0024] Specifically, the collection module comprises real-time acquisition of a device list of the data center infrastructure when a system start instruction is triggered, and the device list records structured data including device type, device model, device quantity, device installation location, commissioning date, rated parameter, maintenance status, and historical fault code; dynamic factors are extracted according to the type of the data center, and the dynamic factors include SLA weight and operation and maintenance mode.

[0025] The device list is the core input of the system, and contains structured data such as Figure 2 The structured data is as follows: For example, the device list input of a certain data center chiller unit is as follows: "Device Type": "Chiller Unit", "Device Model": "Trane CVHG-123", "Device Quantity": 6, "Commissioning Date": "2018-06-01", "Rated Parameter": {"Cooling Capacity": "2000 RT"}, "Maintenance Status": "Manufacturer Maintenance (to 2025)", "Fault History": ["2023-03-12: E102", "2023-08-21: Pump_Fail"].

[0026] Specifically, the work hour prediction module comprises a preprocessing unit, a similarity unit, and a prediction unit: The preprocessing unit is configured to extract corresponding device standard life and basic work hours from a preset knowledge base according to the device model, and calculate the health coefficient corresponding to the device model. The similarity unit is configured to extract the device standard life and basic work hours corresponding to the device model of the device of the similar device type when the preset knowledge base does not contain the device model. The prediction unit is configured to construct a man-hour prediction model, input device model, device quantity, basic man-hour and health coefficient, and output total man-hour, and the prediction formula is:

[0027] wherein, represents total man-hour, N represents the number of device models, represents the number of devices of the i-th device model, represents the basic man-hour of the i-th device model, represents the health coefficient of the i-th device model. The device quantity refers to the number of devices of the same type and the same model.

[0028] According to the device model "Teling CVHG-123", the standard service life and basic man-hour are extracted from the knowledge base: 16 hours / month / unit, According to the commissioning date, the actual service life is calculated = 2025-2018 = 7 years; According to the actual service life, the service life coefficient is calculated = 1+0.02×(7-5)=1.04; According to the number of failures in 12 months, 2 times, the failure coefficient is calculated = 1+0.1×2=1.2; According to the service life coefficient and the failure coefficient, the health score of the device "Teling CVHG-123" is calculated = 1.04×1.2=1.248.

[0029] Specifically, the specific implementation process of calculating the health coefficient corresponding to the device model includes: According to the commissioning date of the device model, the actual service life of the device model is obtained, and the service life coefficient of the device model is calculated , the calculation formula is:

[0030] wherein, Y represents the actual service life of the device model, represents the health coefficient increment of the device model; the health coefficient increment is determined by a preset health coefficient increment template, and the health coefficient increment template includes the mapping rule of the ratio of the actual service life of the device model to the standard service life and the health coefficient increment, such as Figure 3 ; The number of failures of the device model in a time period is counted, and the failure coefficient of the device model is calculated , the calculation formula is:

[0031] wherein, R represents the number of failures of the device model in a time period, Indicates the preset coefficient; Based on the operating environment of the equipment, the environmental coefficient of the equipment model is extracted from a preset environmental coefficient table; The health score of the equipment model is obtained by multiplying the age coefficient, failure coefficient, and environmental coefficient of the equipment model. Based on the maintenance status, installation location, and quantity of the equipment, the maintenance coefficient, location coefficient, and cluster coefficient are extracted from a preset dynamic correction table. The health coefficient corresponding to the equipment model is the product of the health score, maintenance coefficient, location coefficient, and cluster coefficient of the equipment model.

[0032] Specifically, the selection process for devices of the similar device types includes: Obtain the device type, rated parameters, manufacturer, complexity coefficient, and key technologies of the device; calculate the similarity between the device and different device models in a pre-set knowledge base; the calculation formula is as follows:

[0033]

[0034] in, This indicates the similarity between the stated device and devices of the j-th device model in a pre-set knowledge base. This value indicates whether the data types of two devices are the same. The value is 1 when the data type of the device is the same as that of the j-th device model in the preset knowledge base, and 0 when the data type of the device is different from that of the j-th device model in the preset knowledge base. This indicates a value used to reflect the degree of similarity between the rated parameters of two devices. This represents the reciprocal of the number of key technologies that the device differs from the j-th device model in the pre-set knowledge base, and when the number of key technologies that differ is 0, The value is 1. This value indicates whether the manufacturers of the two devices are the same. The value is 1 when the manufacturer of the device is the same as that of the j-th device model in the pre-set knowledge base, and 0 when the manufacturer of the device is different from that of the j-th device model in the pre-set knowledge base. Let Q represent the complexity coefficient, and let Q represent the number of possible parameters. This represents the parameter value of the xth rated parameter of the equipment. This represents the parameter value of the x-th rated parameter of the j-th equipment model. Indicates weight, such as Figure 4 The matching description, and ; The device models are sorted from highest to lowest similarity to generate a similar sequence, and the device with the first similar model is selected as the device of similar device type.

[0035] Specifically, the configuration module includes: The automation replacement rate (ple) is extracted based on the operation and maintenance model. The automation replacement rate is as follows: Figure 5 Obtain the data center employee hour base SP, calculate the final configured number of employees GH, and use the following formula:

[0036] in, This represents the SLA weight.

[0037] Specifically, the SLA weight is the redundancy amplification factor required by the service level agreement of the data center, with a value range of 1.0 to 1.5.

[0038] Specifically, the preset scheduling system includes inputting the final number of personnel and triggering an instruction to generate an executable personnel configuration scheme.

[0039] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0040] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A data center infrastructure personnel intelligent configuration system, characterized in that, The system includes a data acquisition module, a time prediction module, a configuration module, and a feedback module. The acquisition module is used to obtain a list of devices and extract dynamic factors; The time prediction module is used to predict working hours based on the equipment list; The configuration module is used to configure the number of personnel based on the prediction results; The feedback module is used to output an executable personnel configuration plan through a pre-set scheduling system.

2. The intelligent personnel configuration system for data center infrastructure according to claim 1, characterized in that, The acquisition module includes acquiring a real-time equipment list of the data center infrastructure when a system startup command is triggered. The equipment list records structured data including equipment type, equipment model, equipment quantity, equipment installation location, commissioning date, rated parameters, maintenance status, and historical fault codes. Dynamic factors are extracted based on the type of data center. The dynamic factors include SLA weights and operation and maintenance modes.

3. The intelligent personnel configuration system for data center infrastructure according to claim 2, characterized in that, The time prediction module includes a preprocessing unit, a similarity unit, and a prediction unit: The preprocessing unit is used to extract the corresponding standard service life and basic working hours of the equipment from a preset knowledge base according to the equipment model, and calculate the health coefficient corresponding to the equipment model. The similarity unit is used to extract the standard service life and basic working hours of equipment models of similar equipment types when the equipment model does not exist in the preset knowledge base. The prediction unit is used to construct a working hour prediction model. It takes as input equipment model, equipment quantity, basic working hours, and health coefficient, and outputs total working hours. The prediction formula is: ； in, N represents the total working hours, and N represents the number of equipment models. This represents the number of devices of the i-th type. This represents the basic working hours for the i-th equipment model. This represents the health coefficient of the i-th equipment model.

4. The intelligent personnel configuration system for data center infrastructure according to claim 3, characterized in that, The specific implementation process for calculating the health coefficient corresponding to the equipment model includes: Based on the commissioning date of the equipment model, obtain the actual operating years of the equipment model, and calculate the service life coefficient of the equipment model. The calculation formula is: ； Where Y represents the actual operating years of the equipment model. This represents the health coefficient increment of the equipment model; the health coefficient increment is determined by a preset health coefficient increment template, which includes the ratio of the actual operating years of the equipment model to the standard operating years and the mapping rules for the health coefficient increment; Statistically count the number of failures of the specified equipment model within a certain time period, and calculate the failure coefficient of the specified equipment model. The calculation formula is: ； Where R represents the number of failures of the device model within a time period. Indicates the preset coefficient; Based on the operating environment of the equipment, the environmental coefficient of the equipment model is extracted from a preset environmental coefficient table; The health score of the equipment model is obtained by multiplying the age coefficient, failure coefficient, and environmental coefficient of the equipment model. Based on the maintenance status, installation location, and quantity of the equipment, the maintenance coefficient, location coefficient, and cluster coefficient are extracted from a preset dynamic correction table. The health coefficient corresponding to the equipment model is the product of the health score, maintenance coefficient, location coefficient, and cluster coefficient of the equipment model.

5. The intelligent personnel configuration system for data center infrastructure according to claim 3, characterized in that, The selection process for devices of similar device types includes: Obtain the device type, rated parameters, manufacturer, complexity coefficient, and key technologies of the device; calculate the similarity between the device and different device models in a pre-set knowledge base; the calculation formula is as follows: ； ； in, This indicates the similarity between the stated device and devices of the j-th device model in a pre-set knowledge base. This indicates a value used to reflect whether the data types of two devices are the same. This indicates a value used to reflect the degree of similarity between the rated parameters of two devices. This represents the reciprocal of the number of key technologies that the device differs from the j-th device model in the pre-set knowledge base, and when the number of key technologies that differ is 0, The value is 1. This indicates a value used to reflect whether two devices are manufactured by the same company. Let Q represent the complexity coefficient, and let Q represent the number of possible parameters. This represents the parameter value of the xth rated parameter of the equipment. This represents the parameter value of the x-th rated parameter of the j-th equipment model. , , , , Indicates weight; The device models are sorted from highest to lowest similarity to generate a similar sequence, and the device with the first similar model is selected as the device of similar device type.

6. The intelligent personnel configuration system for data center infrastructure according to claim 3, characterized in that, The configuration module includes: Based on the operation and maintenance model, extract the automation replacement rate (ple), obtain the data center man-hour base (SP), and calculate the final number of personnel (GH) using the following formula: ； in, This represents the SLA weight.

7. The intelligent personnel configuration system for data center infrastructure according to claim 6, characterized in that: The SLA weight is the redundancy amplification factor required by the data center's service level agreement.

8. The intelligent personnel configuration system for data center infrastructure according to claim 1, characterized in that: The preset scheduling system includes inputting the final number of personnel and triggering an instruction to generate an executable personnel configuration plan.

Images