Server hardware system evaluation method, device, equipment and medium

Abstract

The invention provides a server hardware system evaluation method and device, equipment and a medium, and the method comprises the steps: preparing n to-be-evaluated servers with the same brand and model and the same operation duration, collecting the historical fault data, calculating the fault occurrence frequency and probability, evaluating the severity of the influence of the fault on the system reliability, and calculating the reliability of the system according to the severity. Calculating the uncertainty and the weighted information amount of the weighted information entropy for representing the hardware fault, and finally converting the weighted information entropy into an evaluation score based on a mapping function; according to the invention, reliability evaluation and comparison of universal hardware systems suitable for various different brands and models of servers can be realized.

Description

Technical Field

[0001] The present application relates to the technical field of server evaluation, and in particular to a server hardware system evaluation method, device, electronic device and storage medium. Background Art

[0002] In today's server market, there are many manufacturers, various brands and models emerge in an endless stream, the manufacturing threshold is low, the quality varies, and the reliability of server hardware is difficult to effectively compare. For enterprises, servers are the core components of their information systems. Reliable server hardware can ensure business continuity and data security. Therefore, the evaluation of server hardware reliability is crucial.

[0003] However, there are currently few systematic methods to use server hardware failures to conduct relevant evaluations on the reliability of server hardware systems. This server selection process, which lacks detailed scientific basis, exposes many companies to potential operational risks and economic losses. The introduction of a reliability evaluation method based on hardware failure analysis can help companies more accurately select high-quality server equipment, thereby improving the stability and efficiency of overall business operations. Summary of the invention

[0004] In view of this, the embodiments of the present application provide a server hardware system evaluation method, apparatus, device and medium, which can realize general hardware system reliability evaluation and comparison applicable to various brands and models of servers.

[0005] The technical solution of the embodiment of the present application is implemented as follows:

[0006] In a first aspect, an embodiment of the present application provides a server hardware system evaluation method, the method comprising:

[0007] Prepare server evaluation resources, obtain n servers to be evaluated with the same brand, model and running time, collect historical fault data of the n servers to be evaluated within the running time T, and calculate the average number of various faults that occur on each server to be evaluated;

[0008] Convert the average number of failures of each type on each server to be evaluated into the probability of occurrence of that type of failure;

[0009] Based on the preset evaluation criteria, the severity of the impact of various failures that occurred in the server to be evaluated in the past on the system reliability is evaluated and processed to obtain a comprehensive severity score;

[0010] Calculating the weighted information entropy of the server to be evaluated based on the occurrence probability and the comprehensive score of the severity, wherein the weighted information entropy is used to characterize the uncertainty of the occurrence of a hardware failure of the server to be evaluated and to obtain the weighted information amount of the probability distribution of the hardware failure of the server to be evaluated, and the weight item is determined based on the occurrence probability and the comprehensive score of the severity;

[0011] The weighted information entropy is converted into an evaluation score result based on a mapping function.

[0012] In a second aspect, an embodiment of the present application further provides a server hardware system evaluation device, the device comprising:

[0013] The acquisition module is used to prepare server evaluation resources, obtain n servers to be evaluated with the same brand and model and the same operating time, collect the historical fault data of the n servers to be evaluated within the operating time T, and calculate the average number of occurrences of various types of faults of each server to be evaluated;

[0014] A conversion module, used to convert the average number of occurrences of each type of failure of each server to be evaluated into the probability of occurrence of the failure;

[0015] An evaluation module is used to evaluate the severity of the impact of various failures that occurred in the server to be evaluated over a historical period on the system reliability based on a preset evaluation standard, and obtain a comprehensive severity score;

[0016] A calculation module, used to calculate the weighted information entropy of the server to be evaluated based on the occurrence probability and the comprehensive score of the severity, wherein the weighted information entropy is used to characterize the uncertainty of the occurrence of hardware failure of the server to be evaluated and to obtain the weighted information amount of the probability distribution of the hardware failure of the server to be evaluated, and the weight item is determined based on the occurrence probability and the comprehensive score of the severity;

[0017] The conversion module is used to convert the weighted information entropy into an evaluation score result based on a mapping function.

[0018] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to execute the server hardware system evaluation method described in any one of the first aspects.

[0019] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the server hardware system evaluation method described in any one of the first aspects is executed.

[0020] The embodiments of the present application have the following beneficial effects:

[0021] (1) In view of the current situation in which there are many manufacturers in the server market, various brands and models are emerging in an endless stream, the quality varies greatly, and the reliability of server hardware is difficult to effectively compare, the embodiments of the present application propose a comprehensive and systematic method to use the hardware failures that occur in the server itself to conduct relevant reliability evaluation of the server hardware system.

[0022] (2) The embodiment of the present application uses the concept of information entropy based on information theory to comprehensively quantify the uncertainty of the frequency and type of server system failures. Traditional server reliability evaluation methods usually rely on fixed indicators, such as mean time between failures (MTBF) and failure rate. Although these indicators can provide a certain reference value, they often ignore the complexity and diversity of various types of failure events in the system. The evaluation method based on information entropy of the present invention comprehensively focuses on various hardware failures of all sizes that may occur in the server system. By calculating the entropy value of the failure event, it reflects the overall disorder of the server system in a specific time period, and has a higher reliability.

[0023] (3) The evaluation method based on information entropy adopted in the embodiment of the present application has higher applicability and versatility. Since information entropy does not depend on a specific hardware or software platform, it can be applied to servers of various brands and models. This allows enterprises to make comparisons based on unified standards when selecting servers, thereby making more informed purchasing decisions. In contrast, other evaluation methods often need to be customized for different systems, making it difficult to achieve unified evaluation across platforms.

[0024] (4) The embodiment of the present application adopts an evaluation method based on information entropy. The evaluation process only requires the hardware failure data of the server itself, does not rely on external tools, and is not affected by the update iteration of server hardware and software. The evaluation method has a high degree of stability and independence. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 is a flowchart of steps S101-S105 provided in an embodiment of the present application;

[0027] Figure 2 It is a schematic diagram of the server hardware system evaluation provided in the embodiment of the present application;

[0028] Figure 3 It is a structural diagram of a server hardware system evaluation device provided in an embodiment of the present application;

[0029] Figure 4 It is a schematic diagram of the composition structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0030] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the present application, or remove one or more operations from the flowchart.

[0031] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0032] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0033] In the following description, the terms "first\second\third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0034] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0036] See also Figure 1 and Figure 2 , Figure 1 1 is a flow chart of steps S101-S105 of the server hardware system evaluation method provided in an embodiment of the present application. Figure 2 This is a schematic diagram of the server hardware system evaluation provided by the embodiment of the present application, which will be combined with Figure 1 Steps S101-S105 and Figure 2 Provide explanation.

[0037] In step S101, server evaluation resources are prepared, n servers to be evaluated with the same brand and model and the same operating time are obtained, historical fault data of the n servers to be evaluated within the operating time T are collected, and the average number of occurrences of various types of faults of each server to be evaluated is calculated.

[0038] Here, we first need to prepare n servers to be evaluated that are of the same brand, model, and running time. This step ensures the consistency of the basic conditions of the evaluation objects, thereby improving the accuracy and comparability of the evaluation results. Next, collect the historical failure data of these servers within the running time T. These data should include the number of occurrences of various types of failures for subsequent statistical analysis.

[0039] In some embodiments, the server evaluation resource is prepared, n servers to be evaluated are obtained with the same brand model and the same running time, historical fault data of the n servers to be evaluated within the running time T are collected, and the average number of occurrences of various types of faults of each server to be evaluated is calculated, including:

[0040] Count the running time T;

[0041] All possible hardware failures of the server to be evaluated are expressed as a failure set:

[0042] {X1,X2,X3,X4,x5,x6,x7,x8,x9,x 10 ,X 11 ,x 12 ,x 13 ,x 14 ,x 15}

[0043] Among them, x1 indicates a motherboard failure, x2 indicates a CPU failure, x3 indicates a power failure, x4 indicates a memory failure, x5 indicates a hard disk failure, x6 indicates a network card failure, x7 indicates a Raid card failure, x8 indicates an HBA card failure, x9 indicates an expansion card failure, x10 indicates a motherboard failure, x11 indicates a CPU failure, x12 indicates a power failure, x13 indicates a memory failure, x14 indicates a hard disk failure, x15 indicates a network card failure, x16 indicates a network card failure, x17 indicates a Raid card failure, x18 indicates a HBA card failure, x19 indicates an expansion card failure, x20 indicates a motherboard failure, x21 indicates a power failure, x22 indicates a power failure, x23 indicates a power failure 10 Indicates PCIe failure, x 11 Indicates a fan / heat sink failure, x 12 Indicates compatibility failure, x 13 Indicates a management module failure, x 14 Indicates that the indicator light is faulty, x 15 Indicates other faults;

[0044] Statistics show that n servers to be evaluated have failed during the running time T. i Count i .

[0045] Calculate the average number of failures x per server to be evaluated during the running time T i Number of occurrences N i , the formula is:

[0046]

[0047] In step S102, the average number of occurrences of each type of failure of each server to be evaluated is converted into the probability of occurrence of the failure of that type.

[0048] Here, the average number of failures of each server to be evaluated is converted into the probability of occurrence of that type of failure. This step helps to convert the original failure number data into a more comparable probability value, which is convenient for subsequent analysis and calculation.

[0049] In some embodiments, converting the average number of occurrences of each type of failure of each server to be evaluated into the probability of occurrence of the failure includes:

[0050] Statistics of failures whose probability of occurrence depends only on the time interval and is independent of past failures;

[0051] According to the statistical data, the failure rate of each type of failure is calculated for the server to be evaluated. The formula is:

[0052]

[0053] Among them, λ i is the failure rate, in times / unit time, N i The number of failures that occurred during the statistical time period;

[0054] Approximately calculate the failure x of the server to be evaluated through exponential distribution i The probability of occurrence is as follows:

[0055]

[0056] Among them, t is the unit time, P(x i ) is fault x i The probability of occurrence within unit time t.

[0057] In step S103, the severity of the impact of various types of failures that occurred in the server to be evaluated in the past on the system reliability is evaluated based on a preset evaluation standard to obtain a comprehensive severity score.

[0058] Here, based on the preset evaluation criteria, the severity of the impact of various types of failures that occurred in the evaluation server over the historical period on the system reliability is evaluated and processed to obtain a comprehensive severity score. This step takes into account the different impacts of different failures on system reliability, making the evaluation results more comprehensive and accurate.

[0059] In some embodiments, the severity of the impact of various types of failures that occurred in the server to be evaluated in the past time on the system reliability is evaluated based on the preset evaluation criteria to obtain a comprehensive severity score, including:

[0060] The evaluation criteria are set based on at least one of system availability, impact scope, repair time, fault location difficulty, data security, redundancy mechanism, and cost impact; wherein, the system availability represents the impact of the fault on the normal operation of the system and service availability; the impact scope represents the number of services or users affected by the fault; the repair time represents the time required to repair the fault; the fault location difficulty represents the ease of locating which component caused the fault after the fault occurs; the data security represents the impact of the fault on data integrity and security; the redundancy mechanism represents whether the server hardware that caused the fault can be designed with a redundancy mechanism to achieve high availability and fault tolerance of the system; the cost impact represents the economic losses caused by the fault repair, including the material and labor costs of replacing parts and the losses caused by business interruption;

[0061] Set quantitative indicators for each of the evaluation criteria;

[0062] For details, please refer to Table 1, which is a reference table for quantitative indicators:

[0063]

[0064]

[0065] Based on the quantitative indicators, the impact of each type of hardware failure on system availability, the scope of impact, the length of repair time, the difficulty of fault location, the data security issues caused, the redundancy mechanism and the cost impact are scored. The scores of each item are accumulated to obtain the comprehensive score of each type of hardware failure. i .

[0066] In step S104, the weighted information entropy of the server to be evaluated is calculated based on the occurrence probability and the comprehensive score of the severity, wherein the weighted information entropy is used to characterize the uncertainty of the hardware failure of the server to be evaluated and to obtain the weighted information amount of the hardware failure probability distribution of the server to be evaluated, and the weight item is determined based on the occurrence probability and the comprehensive score of the severity.

[0067] Here, based on the comprehensive score of the probability and severity of failure, the weighted information entropy of the server to be evaluated is calculated. Weighted information entropy is a comprehensive indicator used to characterize the uncertainty of hardware failures in the server and to obtain the weighted information of the probability distribution of hardware failures in the server. The weight item is determined based on the comprehensive score of the probability and severity of failures, ensuring the objectivity and accuracy of the evaluation results.

[0068] In some embodiments, the calculating the weighted information entropy of the server to be evaluated based on the occurrence probability and the comprehensive score of severity includes:

[0069] The event that the server to be evaluated may fail is represented as a discrete random variable X, and the possible values ​​of the variable are {x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15}, the corresponding probability distribution is:

[0070] P(X)={p(x1),p(x2),p(x3),p(x4),p(x5),p(x6),p(x7),p(x8),

[0071] p(x9),p(x 10 ),p(x 11 ),p(x 12 ),p(x 13 ),p(x 14 ),p(x 15 )};

[0072] Calculate the weighted information entropy of the server to be evaluated, the formula is:

[0073]

[0074] Where m is the total number of fault categories, i is the i-th type of fault x i , 1≤i≤m, H w (X) is the weighted information entropy of the discrete random variable X, and the weight term w 1,i For various faults x i The comprehensive fault score, w 1,i >0, used to determine fault x i The severity of the impact on system reliability, weighted by w 2,i For various faults x i The frequency of occurrence, 0≤w 2,i ≤1, used to determine fault x i is an occasional fault or a frequent fault; p(x i ) indicates that the random variable X takes the value x i The probability that the server to be evaluated has x i The probability of failure; weight term w 2,i And the probability p(x i ) are related to the frequency of failure, p(x i ) is the probability of actual occurrence, p(x i ) Based on data statistics, the weight w 2,i For failures during evaluation x i The importance of the frequency of occurrence is set based on business needs and opinion information; the logarithm of the probability is taken a p(x i ) is the amount of information measured, a is the base of the logarithm, which is 2 and represents the information entropy in bits.

[0075] In step S105, the weighted information entropy is converted into an evaluation score result based on a mapping function.

[0076] Here, the weighted information entropy is converted into the evaluation score result based on the mapping function. This step converts the complex weighted information entropy value into an intuitive evaluation score, which makes it easier for users to understand and compare the evaluation results of different servers.

[0077] In some embodiments, the value range of the weighted information entropy is set; the weight term w in the weighted information entropy formula 1,i 、w 2,i is a value greater than 0, p(x i ) indicates that the random variable X takes the value x i The probability, probability p(x i ) is between 0 and 1, and the entropy H w (X) is non-negative; when the server system does not have any failure, the probability p(x i )=0, then entropy H w(X) obtains its minimum value H min (X) = 0; the maximum value of the weighted information entropy occurs when the probability of the event is uniformly distributed. The calculation of the maximum value is based on the number of events and the distribution of weights. When the weights are not equal, the maximum value of the weighted information entropy can be calculated by an optimization algorithm, which includes the Lagrange multiplier method. In actual applications, the probability of various hardware failures of servers is not uniformly distributed. At the same time, since the probability of server hardware failure is usually very low and is a low-probability event, the weighted information entropy values ​​calculated for servers of different models are all decimals close to 0. Therefore, calculating the actual maximum value of the weighted information entropy has no practical significance for the reliability evaluation of the server hardware system.

[0078] Since the probability of server hardware failure is usually very low and is a low-probability event, the weighted information entropy values ​​calculated for servers of different models are all decimals close to 0, resulting in unclear differences in failure rates between different servers, which will affect the effectiveness of subsequent steps. Therefore, when designing the mapping function, the following mapping rules should be considered: (1) The mapping function can amplify smaller weighted information entropy differences into larger score result differences; (2) When the weighted information entropy H w (X) obtains its minimum value H min When (X) = 0, the entropy value is the lowest, indicating that the server system to be evaluated is extremely stable, will not have any failures, has the highest reliability, and the mapping result is full score (100 points); (3) When the entropy H w (X) The larger the value, the more chaotic the server system to be evaluated is during this period, the more prone to hardware failure, the lower the reliability of the system, and the closer the mapping result is to 0; (4) All mapping results are within the range of 0 to 100. Based on the above mapping rules, a feasible mapping function design is:

[0079] s=100·e -lh ;

[0080] Among them, h is the weighted information entropy value before mapping, s is the score result value after mapping, k is the difference amplification coefficient, k>0;

[0081] Based on the mapping function, the weighted information entropy is converted into an intuitive evaluation score of the server to obtain the final evaluation score result; the conversion process is:

[0082]

[0083] Below, the embodiments of the present application will be fully explained in combination with data from actual applications:

[0084] S1. Prepare server evaluation resources. In this embodiment, n servers to be evaluated are equipped with a certain brand A model and the same running time, where n=3. The historical fault data of the n servers within the running time T are collected. The historical server faults include motherboard faults, CPU faults, power supply faults, memory faults, hard disk faults, network card faults, Raid card faults, HBA card faults, expansion card faults, PCIe faults, fan / radiator faults, compatibility faults, management module faults, indicator light faults and other faults. Calculate the average number of times each type of fault occurs for each server model to be evaluated.

[0085] Step S101, determine the statistical operation time T of the server model to be evaluated. In this embodiment, since the probability of server hardware failure is usually very low and is a low-probability event, the month is used as the basic unit, and T=1 year=12 months is set.

[0086] Step S102: All possible hardware failures of the server are represented as a failure set:

[0087] {x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15}

[0088] Among them, x1 indicates a motherboard failure, x2 indicates a CPU failure, x3 indicates a power failure, x4 indicates a memory failure, x5 indicates a hard disk failure, x6 indicates a network card failure, x7 indicates a Raid card failure, x8 indicates an HBA card failure, x9 indicates an expansion card failure, x10 indicates a motherboard failure, x11 indicates a CPU failure, x12 indicates a power failure, x13 indicates a memory failure, x14 indicates a hard disk failure, x15 indicates a network card failure, x16 indicates a network card failure, x17 indicates a Raid card failure, x18 indicates a HBA card failure, x19 indicates an expansion card failure, x20 indicates a motherboard failure, x21 indicates a power failure, x22 indicates a power failure, x23 indicates a power failure 10 Indicates PCIe failure, x 11 Indicates a fan / heat sink failure, x 12 Indicates compatibility failure, x 13 Indicates a management module failure, x 14 Indicates that the indicator light is faulty, x 15 Indicates other faults.

[0089] Step S103: Count the number of failures of n=200 servers of the model to be evaluated within a period of T=1 year=12 months. i Count i In this embodiment, the statistical results are: Count1 = 2, Count2 = 2, Count3 = 0, Count4 = 3, Count5 = 5, Count6 ​​= 0, Count7 = 0, Count8 = 0, Count9 = 0, Count 10 =0, Count 11=1, Count 12 =0, Count 13 =0, Count 14 =0, Count 15 =1.

[0090] Step S104: Calculate the running time T = 1 year = the average number of failures per server of the model to be evaluated x in 12 months i Number of occurrences N i , the specific formula is:

[0091]

[0092] In this embodiment, the calculated results are: N1=0.01, N2=0.01, N3=0, N4=0.015, N5=0.025, N6=0, N7=0, N8=0, N9=0, N 10 =0,N 11 =0.005, N 12 =0,N 13 =0,N 14 =0,N 15 =0.005.

[0093] S2. Convert the average number of failures of each type of server to be evaluated into the probability of occurrence of that type of failure.

[0094] Step S201: determine that the time of occurrence of the hardware failure counted by the server model to be evaluated is independent of the time of occurrence of the previous hardware failure, that is, the probability of failure only depends on the time interval and has nothing to do with the failures that occurred in the past.

[0095] Step S202: Calculate the failure rate of each type of failure for the server model to be evaluated based on the collected data. The formula is:

[0096]

[0097] Among them, λ i is the failure rate (times / unit time), N i is the number of failures that occurred during the statistical time period, and T is the running time of the statistical time period. In this embodiment, the month is used as the time unit, and the calculated result is:

[0098]

[0099]

[0100] Step S203: Based on the collected data, use exponential distribution to approximate the number of server failures x i The failure probability is:

[0101]

[0102] Among them, t is the unit time, P(x i ) is fault x i The probability of failure occurring within a unit time t. In this embodiment, t=12 months is used as the unit. The calculated result is:

[0103]

[0104] S3. Evaluate the severity of the impact of various types of failures that have occurred in the historical period of the server model to be evaluated on the system reliability, including setting evaluation standards, scoring according to the standards, and accumulating the comprehensive scores.

[0105] Step S301, setting various fault severity assessment standards for the server model to be evaluated, mainly including: system availability: that is, the impact of the fault on the normal operation of the system and service availability; impact range, that is, the number of services or users affected by the fault; repair time, that is, the time required to repair the fault; data security, that is, the impact of the fault on data integrity and security; fault location difficulty, that is, after the fault occurs, the difficulty of locating which component caused the fault; redundancy mechanism, that is, whether the server hardware that caused the fault can be designed with a redundant mechanism to achieve high availability and fault tolerance of the system; cost impact, that is, fault repair and the economic losses caused, including the material and labor costs of replacing parts and the losses caused by business interruption.

[0106] Step S302: quantify the severity evaluation criteria of various types of faults of the server model to be evaluated, and set specific quantitative indicators for each evaluation indicator. The quantitative indicator design method is shown in Table 1.

[0107] Step S303: Based on the severity evaluation criteria for various types of faults of the server to be evaluated, score the impact of each type of hardware fault on system availability, the size of the impact range, the length of repair time, the difficulty of fault location, the data security issues caused, the redundancy mechanism, and the cost impact. The scores of each item are accumulated to obtain a comprehensive score for each type of hardware fault. i In this embodiment, the specific scoring rules for the quantitative indicators are as follows:

[0108]

[0109]

[0110] S4. Based on the probability of occurrence of various types of failures in the server model to be evaluated and the comprehensive score of the severity of the impact on system reliability, calculate the weighted information entropy of the server model to be evaluated to measure the uncertainty of hardware failures in the server model and obtain the weighted information volume of the probability distribution of hardware failures of the server, where the weight term is the comprehensive score of the severity of the impact of various types of failures on system reliability and the frequency of occurrence.

[0111] Step S401: The event that the server model to be evaluated may fail is represented as a discrete random variable X. The possible values ​​of the variable are {x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15}, the corresponding probability distribution is:

[0112] P(X)={p(x1),p(x2),p(x3),p(x4),p(x5),p(x6),p(x7),p(x8),

[0113] p(x9),p(x 10 ),p(x 11 ),p(x 12 ),p(x 13 ),p(x 14 ),p(x 15 )}

[0114] Step S402: Calculate the weighted information entropy of the server model to be evaluated. The specific formula is:

[0115]

[0116] Where m is the total number of fault categories, i is the i-th type of fault x i (1≤i≤m), H w (X) is the weighted information entropy of the discrete random variable X. In information theory, entropy can measure the uncertainty of information or the average measure of information, and characterize the degree of disorder or chaos of the system. High entropy means that the message has higher uncertainty or more information, that is, the system is more chaotic. The weight term w 1,i For various faults x i The comprehensive fault score, w 1,i >0, used to measure the fault x i The severity of the impact on system reliability, weighted by w 2,i For various faults x i The frequency of occurrence, 0≤w 2,i ≤1, used to measure fault xi It is an occasional fault or a frequent fault. p(x i ) indicates that the random variable X takes the value x i The probability that the server model to be evaluated will have x i The probability of failure. The weight term w 2,i And the probability p(x i ) are related to the frequency of failure, but p(x i ) is the actual probability of occurrence, which is obtained based on data statistics, and the weight w 2,i Reflects the failure x during the evaluation process i The importance of the frequency of occurrence is set manually based on specific business needs and expert opinions. a p(x i ) is the amount of information measured, a is the base of the logarithm, usually taken as 2, representing the information entropy in bits. In this embodiment, m=15, the weight term w 1,i The value is the comprehensive fault score Score obtained in step S303 i , that is, w 1,i =Score i The weight term w 2,i The value is the λ obtained in S202 i , that is, w 2,i =λ i (0≤w 2,i ≤1), a is the base of the logarithm, which is 2, that is, a=2, and the calculated result is:

[0117]

[0118] S5. Use a mapping function to convert the calculated weighted information entropy of the server model to be evaluated into an intuitive evaluation score result of the server.

[0119] Step S501: Calculate the value range of weighted information entropy. The weight term w in the weighted information entropy formula 1,i 、w 2,i is a value greater than 0, p(x i ) indicates that the random variable X takes the value x i The probability of i ) is between 0 and 1, and the characteristics of the logarithmic function make the logarithmic value negative, so the entropy H w (X) is always non-negative. When the server system does not have any failures, the probability p(x i )=0, then entropy H w (X) obtains its minimum value H min(X) = 0. The maximum value of weighted information entropy occurs when the probability of events is uniformly distributed. The calculation of the maximum value depends on the number of events and the distribution of weights. When the weights are unequal, the maximum value of weighted information entropy can be calculated by optimization algorithms. In theory, specific optimization problems can be solved by mathematical tools such as the Lagrange multiplier method. However, in practical applications, the probability of various hardware failures of servers is not uniformly distributed. At the same time, since the probability of server hardware failure is usually very low and belongs to a low-probability event, the weighted information entropy values ​​calculated for servers of different models are all decimals close to 0. Therefore, calculating the actual maximum value of weighted information entropy has no practical significance for the reliability evaluation of server hardware systems.

[0120] Step S502: Design a mapping function from weighted information entropy to score result. Since the probability of server hardware failure is usually very low and is a low-probability event, the weighted information entropy values ​​calculated for servers of different models are all decimals close to 0, resulting in unclear differences in failure rates between different servers, which will affect the effectiveness of subsequent steps. Therefore, when designing a mapping function, the following mapping rules need to be considered: (1) The mapping function can amplify a smaller weighted information entropy difference into a larger score result difference; (2) When the weighted information entropy H w (X) obtains its minimum value H min When (X) = 0, the entropy value is the lowest, indicating that the server system to be evaluated is extremely stable, will not have any failures, has the highest reliability, and the mapping result is full score (100 points); (3) When the entropy H w (X) The larger the value, the more chaotic the server system to be evaluated is during this period, the more prone to hardware failure, the lower the reliability of the system, and the closer the mapping result is to 0; (4) All mapping results are within the range of 0 to 100. Based on the above mapping rules, a feasible mapping function design is:

[0121] s=100·e -kh

[0122] Wherein, h is the weighted information entropy value before mapping, s is the score result value after mapping, k is the difference amplification coefficient (k>0), and the larger the k value is, the more obvious the difference amplification effect on the weighted information entropy is. In this embodiment, k=50, that is, the mapping function is:

[0123] s=100·e -50h

[0124] Step S503: Use a mapping function to convert the weighted information entropy into a server intuitive evaluation score (with a value of 0-100 points), which is the final reliability evaluation result of the tested model server hardware system:

[0125]

[0126] In the embodiment of the present application, the reliability evaluation result of the server hardware system of a certain brand A model is scored 71.659102 points. Through the above embodiment, the present invention can perform reliability evaluation for the server hardware system based on the collected historical failure data and obtain a comprehensive score result. Furthermore, by comparing the respective score results of servers of different brands and models, the hardware systems of servers of different brands and models can also be compared.

[0127] In summary, the embodiments of the present application have the following beneficial effects:

[0128] (1) In view of the current situation in which there are many manufacturers in the server market, various brands and models are emerging in an endless stream, the quality varies greatly, and the reliability of server hardware is difficult to effectively compare, the embodiments of the present application propose a comprehensive and systematic method to use the hardware failures that occur in the server itself to conduct relevant reliability evaluation of the server hardware system.

[0129] (2) The embodiment of the present application uses the concept of information entropy based on information theory to comprehensively quantify the uncertainty of the frequency and type of server system failures. Traditional server reliability evaluation methods usually rely on fixed indicators, such as mean time between failures (MTBF) and failure rate. Although these indicators can provide a certain reference value, they often ignore the complexity and diversity of various types of failure events in the system. The evaluation method based on information entropy of the present invention comprehensively focuses on various hardware failures of all sizes that may occur in the server system. By calculating the entropy value of the failure event, it reflects the overall disorder of the server system in a specific time period, and has a higher reliability.

[0130] (3) The evaluation method based on information entropy adopted in the embodiment of the present application has higher applicability and versatility. Since information entropy does not depend on a specific hardware or software platform, it can be applied to servers of various brands and models. This allows enterprises to make comparisons based on unified standards when selecting servers, thereby making more informed purchasing decisions. In contrast, other evaluation methods often need to be customized for different systems, making it difficult to achieve unified evaluation across platforms.

[0131] (4) The embodiment of the present application adopts an evaluation method based on information entropy. The evaluation process only requires the hardware failure data of the server itself, does not rely on external tools, and is not affected by the update iteration of server hardware and software. The evaluation method has a high degree of stability and independence.

[0132] Based on the same inventive concept, the embodiment of the present application also provides a server hardware system evaluation device corresponding to the server hardware system evaluation method in the first embodiment. Since the principle of solving the problem by the device in the embodiment of the present application is similar to the above-mentioned server hardware system evaluation method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0133] like Figure 3 As shown, Figure 3 3 is a schematic diagram of the structure of a server hardware system evaluation device 300 provided in an embodiment of the present application. The server hardware system evaluation device 300 includes:

[0134] The acquisition module 301 is used to prepare server evaluation resources, obtain n servers to be evaluated with the same brand and model and the same operating time, collect historical fault data of the n servers to be evaluated within the operating time T, and calculate the average number of occurrences of various types of faults of each server to be evaluated;

[0135] The conversion module 302 is used to convert the average number of occurrences of each type of failure of each server to be evaluated into the probability of occurrence of the failure;

[0136] An evaluation module 303 is used to evaluate the severity of the impact of various types of failures that occurred in the server to be evaluated in the past on the system reliability based on a preset evaluation standard to obtain a comprehensive severity score;

[0137] A calculation module 304 is used to calculate the weighted information entropy of the server to be evaluated based on the occurrence probability and the comprehensive score of the severity, wherein the weighted information entropy is used to characterize the uncertainty of the hardware failure of the server to be evaluated and to obtain the weighted information amount of the probability distribution of the hardware failure of the server to be evaluated, and the weight item is determined based on the occurrence probability and the comprehensive score of the severity;

[0138] The conversion module 305 is used to convert the weighted information entropy into an evaluation score result based on a mapping function.

[0139] Those skilled in the art should understand that Figure 3 The implementation functions of each unit in the server hardware system evaluation device 300 shown can be understood by referring to the relevant description of the aforementioned server hardware system evaluation method. Figure 3 The functions of each unit in the server hardware system evaluation device 300 shown can be implemented by a program running on a processor, or can be implemented by a specific logic circuit.

[0140] In a possible implementation, the acquisition module 301 prepares server evaluation resources, obtains n servers to be evaluated of the same brand and model and the same operating time, collects historical fault data of the n servers to be evaluated within the operating time T, and calculates the average number of occurrences of various types of faults of each server to be evaluated, including:

[0141] Count the running time T;

[0142] All possible hardware failures of the server to be evaluated are expressed as a failure set:

[0143] {x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15}

[0144] Among them, x1 indicates a motherboard failure, x2 indicates a CPU failure, x3 indicates a power failure, x4 indicates a memory failure, x5 indicates a hard disk failure, x6 indicates a network card failure, x7 indicates a Raid card failure, x8 indicates an HBA card failure, x9 indicates an expansion card failure, x10 indicates a motherboard failure, x11 indicates a CPU failure, x12 indicates a power failure, x13 indicates a memory failure, x14 indicates a hard disk failure, x15 indicates a network card failure, x16 indicates a network card failure, x17 indicates a Raid card failure, x18 indicates a HBA card failure, x19 indicates an expansion card failure, x20 indicates a motherboard failure, x21 indicates a power failure, x22 indicates a power failure, x23 indicates a power failure 10 Indicates PCIe failure, x 11 Indicates a fan / heat sink failure, x 12 Indicates compatibility failure, x 13 Indicates a management module failure, x 14 Indicates that the indicator light is faulty, x 15 Indicates other faults;

[0145] Statistics show that n servers to be evaluated have failed during the running time T. i Count i .

[0146] Calculate the average number of failures x per server to be evaluated during the running time T i Number of occurrences N i , the formula is:

[0147]

[0148] In a possible implementation, the conversion module 302 converts the average number of occurrences of each type of failure of each server to be evaluated into the probability of occurrence of the failure, including:

[0149] Statistics of failures whose probability of occurrence depends only on the time interval and is independent of past failures;

[0150] According to the statistical data, the failure rate of each type of failure is calculated for the server to be evaluated. The formula is:

[0151]

[0152] Among them, λ i is the failure rate, in times / unit time, N i The number of failures that occurred during the statistical time period;

[0153] Approximately calculate the failure x of the server to be evaluated through exponential distribution i The probability of occurrence is as follows:

[0154]

[0155] Among them, t is the unit time, P(x i ) is fault x i The probability of occurrence within unit time t.

[0156] In a possible implementation, the evaluation module 303 evaluates the severity of the impact of various types of failures that occurred in the server to be evaluated in the historical period on the system reliability based on the preset evaluation criteria to obtain a comprehensive severity score, including:

[0157] The evaluation criteria are set based on at least one of system availability, impact scope, repair time, fault location difficulty, data security, redundancy mechanism, and cost impact; wherein, the system availability represents the impact of the fault on the normal operation of the system and service availability; the impact scope represents the number of services or users affected by the fault; the repair time represents the time required to repair the fault; the fault location difficulty represents the ease of locating which component caused the fault after the fault occurs; the data security represents the impact of the fault on data integrity and security; the redundancy mechanism represents whether the server hardware that caused the fault can be designed with a redundancy mechanism to achieve high availability and fault tolerance of the system; the cost impact represents the economic losses caused by the fault repair, including the material and labor costs of replacing parts and the losses caused by business interruption;

[0158] Set quantitative indicators for each of the evaluation criteria;

[0159] Based on the quantitative indicators, the impact of each type of hardware failure on system availability, the scope of impact, the length of repair time, the difficulty of fault location, the data security issues caused, the redundancy mechanism and the cost impact are scored. The scores of each item are accumulated to obtain the comprehensive score of each type of hardware failure. i .

[0160] In a possible implementation, the calculation module 304 calculates the weighted information entropy of the server to be evaluated based on the occurrence probability and the comprehensive score of severity, including:

[0161] The event that the server to be evaluated may fail is represented as a discrete random variable X, and the possible values ​​of the variable are {x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15}, the corresponding probability distribution is:

[0162] P(X)={p(x1),p(x2),p(x3),p(x4),p(x5),p(x6),p(x7),p(x8),

[0163] p(x9),p(x 10 ),p(x 11 ),p(x 12 ),p(x 13 ),p(x 14 ),p(x 15 )};

[0164] Calculate the weighted information entropy of the server to be evaluated, the formula is:

[0165]

[0166] Where m is the total number of fault categories, i is the i-th type of fault x i , 1≤i≤m, H w (X) is the weighted information entropy of the discrete random variable X, and the weight term w 1,i For various faults x i The comprehensive fault score, w 1,i >0, used to determine fault x i The severity of the impact on system reliability, weighted by w 2,i For various faults x i The frequency of occurrence, 0≤w 2,i ≤1, used to determine fault x i is an occasional fault or a frequent fault; p(x i ) indicates that the random variable X takes the value x i The probability that the server to be evaluated has x i The probability of failure; weight term w 2,i And the probability p(x i ) are related to the frequency of failure, p(x i ) is the probability of actual occurrence, p(xi ) Based on data statistics, the weight w 2,i For failures during evaluation x i The importance of the frequency of occurrence is set based on business needs and opinion information; the logarithm of the probability is taken a p(x i ) is the amount of information measured, a is the base of the logarithm, which is 2 and represents the information entropy in bits.

[0167] In a possible implementation, the conversion module 305 converts the weighted information entropy into an evaluation score result based on a mapping function, including:

[0168] Set the value range of weighted information entropy; the weight term w in the weighted information entropy formula 1,i 、w 2,i is a value greater than 0, p(x i ) indicates that the random variable X takes the value x i The probability, probability p(x i ) is between 0 and 1, and the entropy H w (X) is non-negative; when the server system does not have any failure, the probability p(x i )=0, then entropy H w (X) obtains its minimum value H min (X) = 0; the maximum value of the weighted information entropy occurs when the probability of the event is uniformly distributed. The calculation of the maximum value is determined based on the number of events and the distribution of weights. When the weights are not equal, the maximum value of the weighted information entropy can be calculated by an optimization algorithm, which includes the Lagrange multiplier method;

[0169] Set the mapping function from weighted information entropy to score result. The mapping function is:

[0170] s=100·e -kh ;

[0171] Among them, h is the weighted information entropy value before mapping, s is the score result value after mapping, k is the difference amplification coefficient, k>0;

[0172] Based on the mapping function, the weighted information entropy is converted into an intuitive evaluation score of the server to obtain the final evaluation score result; the conversion process is:

[0173]

[0174] The server hardware system evaluation device has the following beneficial effects:

[0175] (1) In view of the current situation in which there are many manufacturers in the server market, various brands and models are emerging in an endless stream, the quality varies greatly, and the reliability of server hardware is difficult to effectively compare, the embodiments of the present application propose a comprehensive and systematic method to use the hardware failures that occur in the server itself to conduct relevant reliability evaluation of the server hardware system.

[0176] (2) The embodiment of the present application uses the concept of information entropy based on information theory to comprehensively quantify the uncertainty of the frequency and type of server system failures. Traditional server reliability evaluation methods usually rely on fixed indicators, such as mean time between failures (MTBF) and failure rate. Although these indicators can provide a certain reference value, they often ignore the complexity and diversity of various types of failure events in the system. The evaluation method based on information entropy of the present invention comprehensively focuses on various hardware failures of all sizes that may occur in the server system. By calculating the entropy value of the failure event, it reflects the overall disorder of the server system in a specific time period, and has a higher reliability.

[0177] (3) The evaluation method based on information entropy adopted in the embodiment of the present application has higher applicability and versatility. Since information entropy does not depend on a specific hardware or software platform, it can be applied to servers of various brands and models. This allows enterprises to make comparisons based on unified standards when selecting servers, thereby making more informed purchasing decisions. In contrast, other evaluation methods often need to be customized for different systems, making it difficult to achieve unified evaluation across platforms.

[0178] (4) The embodiment of the present application adopts an evaluation method based on information entropy. The evaluation process only requires the hardware failure data of the server itself, does not rely on external tools, and is not affected by the update iteration of server hardware and software. The evaluation method has a high degree of stability and independence.

[0179] like Figure 4 As shown, Figure 4 The present invention provides a schematic diagram of the structure of an electronic device 400, wherein the electronic device 400 includes:

[0180] A processor 401, a storage medium 402 and a bus 403, wherein the storage medium 402 stores machine-readable instructions executable by the processor 401. When the electronic device 400 is running, the processor 401 communicates with the storage medium 402 via the bus 403, and the processor 401 executes the machine-readable instructions to perform the steps of the server hardware system evaluation method described in the embodiment of the present application.

[0181] In actual application, the components in the electronic device 400 are coupled together via the bus 403. It is understood that the bus 403 is used to realize the connection and communication between these components. In addition to the data bus, the bus 403 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, Figure 4 Various buses are labeled as bus 403.

[0182] The above electronic device has the following beneficial effects:

[0183] (1) In view of the current situation in which there are many manufacturers in the server market, various brands and models are emerging in an endless stream, the quality varies greatly, and the reliability of server hardware is difficult to effectively compare, the embodiments of the present application propose a comprehensive and systematic method to use the hardware failures that occur in the server itself to conduct relevant reliability evaluation of the server hardware system.

[0184] (2) The embodiment of the present application uses the concept of information entropy based on information theory to comprehensively quantify the uncertainty of the frequency and type of server system failures. Traditional server reliability evaluation methods usually rely on fixed indicators, such as mean time between failures (MTBF) and failure rate. Although these indicators can provide a certain reference value, they often ignore the complexity and diversity of various types of failure events in the system. The evaluation method based on information entropy of the present invention comprehensively focuses on various hardware failures of all sizes that may occur in the server system. By calculating the entropy value of the failure event, it reflects the overall disorder of the server system in a specific time period, and has a higher reliability.

[0185] (3) The evaluation method based on information entropy adopted in the embodiment of the present application has higher applicability and versatility. Since information entropy does not depend on a specific hardware or software platform, it can be applied to servers of various brands and models. This allows enterprises to make comparisons based on unified standards when selecting servers, thereby making more informed purchasing decisions. In contrast, other evaluation methods often need to be customized for different systems, making it difficult to achieve unified evaluation across platforms.

[0186] (4) The embodiment of the present application adopts an evaluation method based on information entropy. The evaluation process only requires the hardware failure data of the server itself, does not rely on external tools, and is not affected by the update iteration of server hardware and software. The evaluation method has a high degree of stability and independence.

[0187] The embodiment of the present application also provides a computer-readable storage medium, which stores executable instructions. When the executable instructions are executed by at least one processor 401, the server hardware system evaluation method described in the embodiment of the present application is implemented.

[0188] In some embodiments, the storage medium can be a magnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface storage, an optical disk, or a compact disc read-only memory (CD ROM); it can also be various devices including one or any combination of the above memories.

[0189] In some embodiments, executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0190] As an example, executable instructions may, but do not necessarily, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (for example, files storing one or more modules, subroutines, or code portions).

[0191] By way of example, executable instructions may be deployed to be executed on one computing device, or on multiple computing devices located at one site, or on multiple computing devices distributed across multiple sites and interconnected by a communication network.

[0192] The computer-readable storage medium has the following advantages:

[0193] (1) In view of the current situation in which there are many manufacturers in the server market, various brands and models are emerging in an endless stream, the quality varies greatly, and the reliability of server hardware is difficult to effectively compare, the embodiments of the present application propose a comprehensive and systematic method to use the hardware failures that occur in the server itself to conduct relevant reliability evaluation of the server hardware system.

[0194] (2) The embodiment of the present application uses the concept of information entropy based on information theory to comprehensively quantify the uncertainty of the frequency and type of server system failures. Traditional server reliability evaluation methods usually rely on fixed indicators, such as mean time between failures (MTBF) and failure rate. Although these indicators can provide a certain reference value, they often ignore the complexity and diversity of various types of failure events in the system. The evaluation method based on information entropy of the present invention comprehensively focuses on various hardware failures of all sizes that may occur in the server system. By calculating the entropy value of the failure event, it reflects the overall disorder of the server system in a specific time period, and has a higher reliability.

[0195] (3) The evaluation method based on information entropy adopted in the embodiment of the present application has higher applicability and versatility. Since information entropy does not depend on a specific hardware or software platform, it can be applied to servers of various brands and models. This allows enterprises to make comparisons based on unified standards when selecting servers, thereby making more informed purchasing decisions. In contrast, other evaluation methods often need to be customized for different systems, making it difficult to achieve unified evaluation across platforms.

[0196] (4) The embodiment of the present application adopts an evaluation method based on information entropy. The evaluation process only requires the hardware failure data of the server itself, does not rely on external tools, and is not affected by the update iteration of server hardware and software. The evaluation method has a high degree of stability and independence.

[0197] In the several embodiments provided in the present application, it should be understood that the disclosed methods and electronic devices can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0198] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0199] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0200] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a platform server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.

[0201] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A server hardware system evaluation method, characterized in that: The method comprises: Prepare server evaluation resources, obtain n servers to be evaluated with the same brand, model and running time, collect historical fault data of the n servers to be evaluated within the running time T, and calculate the average number of various faults that occur on each server to be evaluated; Convert the average number of failures of each type on each server to be evaluated into the probability of occurrence of that type of failure; Based on the preset evaluation criteria, the severity of the impact of various failures that occurred in the server to be evaluated in the past on the system reliability is evaluated and processed to obtain a comprehensive severity score; Calculating the weighted information entropy of the server to be evaluated based on the occurrence probability and the comprehensive score of the severity, wherein the weighted information entropy is used to characterize the uncertainty of the occurrence of a hardware failure of the server to be evaluated and to obtain the weighted information amount of the probability distribution of the hardware failure of the server to be evaluated, and the weight item is determined based on the occurrence probability and the comprehensive score of the severity; The weighted information entropy is converted into an evaluation score result based on a mapping function.

2. The method according to claim 1, characterized in that The server evaluation resources are prepared, n servers to be evaluated are obtained with the same brand model and the same running time, historical fault data of the n servers to be evaluated within the running time T are collected, and the average number of occurrences of various types of faults of each server to be evaluated is calculated, including: Count the running time T; All possible hardware failures of the server to be evaluated are expressed as a failure set: {x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15 } Among them, x1 indicates a motherboard failure, x2 indicates a CPU failure, x3 indicates a power failure, x4 indicates a memory failure, x5 indicates a hard disk failure, x6 indicates a network card failure, x7 indicates a Raid card failure, x8 indicates an HBA card failure, x9 indicates an expansion card failure, x10 indicates a motherboard failure, x11 indicates a CPU failure, x12 indicates a power failure, x13 indicates a memory failure, x14 indicates a hard disk failure, x15 indicates a network card failure, x16 indicates a network card failure, x17 indicates a Raid card failure, x18 indicates a HBA card failure, x19 indicates an expansion card failure, x20 indicates a motherboard failure, x21 indicates a power failure, x22 indicates a power failure, x23 indicates a power failure 10 Indicates PCIe failure, x 11 Indicates a fan / heat sink failure, x 12 Indicates compatibility failure, x 13 Indicates a management module failure. 14 Indicates that the indicator light is faulty, x 15 Indicates other faults; Statistics show that n servers to be evaluated have failed during the running time T. i Count i ; Calculate the average number of failures x per server to be evaluated during the running time T i Number of occurrences N i , the formula is:

3. The method according to claim 2, characterized in that The conversion of the average number of occurrences of each type of failure of each server to be evaluated into the probability of occurrence of such failure includes: Statistics of failures whose probability of occurrence depends only on the time interval and is independent of past failures; According to the statistical data, the failure rate of each type of failure is calculated for the server to be evaluated. The formula is: Among them, λ i is the failure rate, in times / unit time, N i The number of failures that occurred during the statistical time period; Approximately calculate the failure x of the server to be evaluated through exponential distribution i The probability of occurrence is as follows: Among them, t is the unit time, P(x i ) is fault x i The probability of occurrence within unit time t.

4. The method according to claim 3, characterized in that The severity of the impact of various failures that occurred in the server to be evaluated in the past on the system reliability is evaluated based on the preset evaluation criteria to obtain a comprehensive severity score, including: The evaluation criteria are set based on at least one of system availability, impact scope, repair time, fault location difficulty, data security, redundancy mechanism, and cost impact; wherein, the system availability represents the impact of the fault on the normal operation of the system and service availability; the impact scope represents the number of services or users affected by the fault; the repair time represents the time required to repair the fault; the fault location difficulty represents the ease of locating which component caused the fault after the fault occurs; the data security represents the impact of the fault on data integrity and security; the redundancy mechanism represents whether the server hardware that caused the fault can be designed with a redundancy mechanism to achieve high availability and fault tolerance of the system; the cost impact represents the economic losses caused by the fault repair, including the material and labor costs of replacing parts and the losses caused by business interruption; Set quantitative indicators for each of the evaluation criteria; Based on the quantitative indicators, the impact of each type of hardware failure on system availability, the scope of impact, the length of repair time, the difficulty of fault location, the data security issues caused, the redundancy mechanism and the cost impact are scored. The scores of each item are accumulated to obtain the comprehensive score of each type of hardware failure. i .

5. The method according to claim 4, characterized in that The calculating of the weighted information entropy of the server to be evaluated based on the occurrence probability and the comprehensive score of the severity includes: The event that the server to be evaluated may fail is represented as a discrete random variable X, and the value of the variable is {x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 , x 11 , x 12 , x 13 , x 14 , x 15 }, the corresponding probability distribution is: P(X)={p(x1), p(x2), p(x3), p(x4), p(x5), p(x6), p(x7), p(x8), p(x9),p(x 10 ),p(x 11 ),p(x 12 ),p(x 13 ),p(x 14 ),p(x 15 )}; Calculate the weighted information entropy of the server to be evaluated, the formula is: Where m is the total number of fault categories, i is the i-th type of fault x i , 1≤i≤m, H w (X) is the weighted information entropy of the discrete random variable X, and the weight term w 1，i For various faults x i The comprehensive fault score, w 1，i >0, used to determine fault x i The severity of the impact on system reliability, weighted by w 2，i For various faults x i The frequency of occurrence, 0≤w 2，i ≤1, used to determine fault x i is an occasional fault or a frequent fault; p(x i ) indicates that the random variable X takes the value x i The probability that the server to be evaluated has x i The probability of failure; weight term w 2，i And the probability p(x i ) are related to the frequency of failures, p(x i ) is the actual probability of occurrence, p(x i ) Based on data statistics, the weight w 2，i For failures during evaluation x i The importance of the frequency of occurrence is set based on business needs and opinion information; the logarithm of the probability is taken a p(x i ) is the amount of information measured, a is the base of the logarithm, which is 2 and represents the information entropy in bits.

6. The method according to claim 5, characterized in that The converting the weighted information entropy into an evaluation score result based on a mapping function includes: Set the value range of weighted information entropy; the weight term w in the weighted information entropy formula 1，i 、w 2,i is a value greater than 0, p(x i ) indicates that the random variable X takes the value x i The probability of p(x i ) is between 0 and 1, and the entropy H w (X) is non-negative; when the server system does not have any failure, the probability p(x i )=0, then entropy H w (X) obtains its minimum value H min (X) = 0; the maximum value of the weighted information entropy occurs when the probability of the event is uniformly distributed. The calculation of the maximum value is determined based on the number of events and the distribution of weights. When the weights are not equal, the maximum value of the weighted information entropy is calculated by an optimization algorithm, which includes a Lagrange multiplier method; Set the mapping function from weighted information entropy to score result. The mapping function is: s=100·e -kh ; Among them, h is the weighted information entropy value before mapping, s is the score result value after mapping, k is the difference amplification coefficient, k ＞ 0; Based on the mapping function, the weighted information entropy is converted into an intuitive evaluation score of the server to obtain the final evaluation score result; the conversion process is:

7. A server hardware system evaluation device, characterized in that: The device comprises: The acquisition module is used to prepare server evaluation resources, obtain n servers to be evaluated with the same brand and model and the same operating time, collect the historical fault data of the n servers to be evaluated within the operating time T, and calculate the average number of occurrences of various types of faults of each server to be evaluated; A conversion module, used to convert the average number of occurrences of each type of failure of each server to be evaluated into the probability of occurrence of the failure; An evaluation module is used to evaluate the severity of the impact of various failures that occurred in the server to be evaluated over a historical period on the system reliability based on a preset evaluation standard, and obtain a comprehensive severity score; A calculation module, used to calculate the weighted information entropy of the server to be evaluated based on the occurrence probability and the comprehensive score of the severity, wherein the weighted information entropy is used to characterize the uncertainty of the occurrence of hardware failure of the server to be evaluated and to obtain the weighted information amount of the probability distribution of the hardware failure of the server to be evaluated, and the weight item is determined based on the occurrence probability and the comprehensive score of the severity; The conversion module is used to convert the weighted information entropy into an evaluation score result based on a mapping function.

8. The device according to claim 7, characterized in that The calculation module calculates the weighted information entropy of the server to be evaluated based on the occurrence probability and the comprehensive score of the severity, including: The event that the server to be evaluated may fail is represented as a discrete random variable X, and the possible values ​​of the variable are {x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 , x 11 , x 12 , x 13 , x 14 , x 15 }, the corresponding probability distribution is: P(X)={p(x1),p(x2),p(x3),p(x4),p(x5),p(x6),p(x7),p(x8),p(x9),p(x 10 ),p(x 11 ),p(x 12 ),p(x 13 ),p(x 14 ),p(x 15 )}; Calculate the weighted information entropy of the server to be evaluated, the formula is: Where m is the total number of fault categories, i is the i-th type of fault x i , 1≤i≤m, H w (X) is the weighted information entropy of the discrete random variable X, and the weight term w 1，i For various faults x i The comprehensive fault score, w 1，i >0, used to determine fault x i The severity of the impact on system reliability, weighted by w 2，i For various faults x i The frequency of occurrence, 0≤w 2，i ≤1, used to determine fault x i is an occasional fault or a frequent fault; p(x i ) indicates that the random variable X takes the value x i The probability that the server to be evaluated has x i The probability of failure; weight term w 2，i And the probability p(x i ) are related to the frequency of failures, p(x i ) is the actual probability of occurrence, p(x i ) Based on data statistics, the weight w 2，i For failures during evaluation x i The importance of the frequency of occurrence is set based on business needs and opinion information; the logarithm of the probability is taken a p(x i ) is the amount of information measured, a is the base of the logarithm, which is 2 and represents the information entropy in bits.

9. An electronic device, characterized in that: include: A processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to execute the server hardware system evaluation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the server hardware system evaluation method according to any one of claims 1 to 7 is executed.

Images