CN107391288A - Server evaluating method and equipment - Google Patents

Server evaluating method and equipment Download PDF

Info

Publication number
CN107391288A
CN107391288A CN201710138866.2A CN201710138866A CN107391288A CN 107391288 A CN107391288 A CN 107391288A CN 201710138866 A CN201710138866 A CN 201710138866A CN 107391288 A CN107391288 A CN 107391288A
Authority
CN
China
Prior art keywords
server
evaluated
manufacturer
value
sampled point
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201710138866.2A
Other languages
Chinese (zh)
Inventor
夏海宏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Publication of CN107391288A publication Critical patent/CN107391288A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/008Reliability or availability analysis

Landscapes

  • Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Debugging And Monitoring (AREA)

Abstract

The purpose of the application is to provide a kind of server evaluating method and equipment,The sampled point that the application passes through each single item performance indications of each evaluation tool of the server of each manufacturer to be evaluated of collection,Wherein,The sampled point collected is the sampled point in preset time,Each sampled point of each single item performance indications of the server of each manufacturer to be evaluated is polymerize,Performance index value and stability indicator value corresponding to respectively obtaining,According to the performance index value and stability indicator value of all performance indications of the server of each manufacturer to be evaluated,It can obtain scoring comprising the whole synthesis for evaluating and testing obtained multinomial performance index and stability by multiple evaluation tools,Not only consider performance indications when realizing and weighing server performance,Consider stability indicator simultaneously,Carried out more comprehensively so as to the performance of the server to each manufacturer to be evaluated,Accurately evaluate and test.

Description

Server evaluating method and equipment
Technical field
The application is related to computer realm, more particularly to a kind of server evaluating method and equipment.
Background technology
The existing reference performance evaluating method to server such as Cloud Server etc., it is limited only to survey by some performance The individual event performance index value that the race of trial work tool or method is got, so as to which contrast evaluation and test can only be carried out to individual event performance indications, comment It is very not comprehensive to survey result, currently existing scheme, server whole synthesis performance can not both be weighed, also can not be to server Stability weighed.Therefore, a kind of clothes that can be evaluated and tested to the combination property and stability of server are needed badly at present Business device evaluating method and equipment.
The content of the invention
The purpose of the application is to provide a kind of server evaluating method and equipment, and solution at present can not be whole to server Body combination property is weighed, the problem of can not also being weighed to the stability of server.
According to the one side of the application, there is provided a kind of server evaluating method, this method include:
The sampled point of each single item performance indications of each evaluation tool of the server of each manufacturer to be evaluated is collected, its In, the sampled point of collection is the sampled point in preset time;
Each sampled point of each single item performance indications of the server of each manufacturer to be evaluated is polymerize, respectively obtained pair The performance index value and stability indicator value answered;
According to the performance index value and stability indicator value of all performance indications of the server of each manufacturer to be evaluated, Obtain the comprehensive grading of the server of each manufacturer to be evaluated.
Further, in the above method, according to the performance of all performance indications of the server of each manufacturer to be evaluated Desired value and stability indicator value, the comprehensive grading of the server of each manufacturer to be evaluated is obtained, including:
The performance index value and stability indicator value of each single item performance indications of the server of each manufacturer to be evaluated are entered Row normalization, obtain corresponding performance indications normalization score and stability indicator normalization score;
According to the performance indications of all performance indications of the server of each manufacturer to be evaluated normalization score and stably Property index normalization score, obtain the comprehensive grading of the server of each manufacturer to be evaluated.
Further, in the above method, the performance of each single item performance indications of the server of each manufacturer to be evaluated is referred to Scale value and stability indicator value are normalized, and obtain corresponding performance indications normalization score and stability indicator normalizes Point, including:
To the performance index value of the server of all manufacturers to be evaluated of each single item performance indications, with all properties desired value The performance index value of the server of middle highest manufacturer to be evaluated is 100 points, to all manufacturers to be evaluated of this performance indications The performance index value of server be normalized, obtain the performance indications hundred of each manufacturer to be evaluated of each single item performance indications Divide to be made and divide;
To the stability indicator value of the server of all manufacturers to be evaluated of each single item performance indications, referred to all stability The stability indicator value of the server of highest manufacturer to be evaluated is 100 points in scale value, to all to be evaluated of this performance indications The stability indicator value of server for surveying manufacturer is normalized, obtain each single item performance indications each manufacturer to be evaluated it is steady Qualitative index hundred-mark system score.
Further, in the above method, according to the performance of all performance indications of the server of each manufacturer to be evaluated Index normalizes score and stability indicator normalization score, obtains the comprehensive grading of the server of each manufacturer to be evaluated, wraps Include:
Calculate all performance indications hundred-mark system scores of each evaluation tool of the server of each manufacturer to be evaluated The average value of average value and all stability indicator hundred-mark system scores, respectively as each survey of the server of the manufacturer to be evaluated Evaluate somebody's work the performance indications hundred-mark system score average and stability indicator hundred-mark system score average of tool;
According to the performance indications hundred-mark system score average of each evaluation tool of the server of each manufacturer to be evaluated and surely Qualitative index hundred-mark system score average, obtains the comprehensive grading of the server of each manufacturer to be evaluated.
Further, in the above method, referred to according to the performance of each evaluation tool of the server of each manufacturer to be evaluated Mark hundred-mark system score average and stability indicator hundred-mark system score average, the synthesis for obtaining the server of the manufacturer to be evaluated are commented Point, including:
According to the performance indications hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated and surely Qualitative index hundred-mark system score average, obtains the performance indications percentage of all evaluation tools of the server of each manufacturer to be evaluated It is made and divides the weighted sum value of average and the weighted sum value of stability indicator hundred-mark system score average;
According to adding for the performance indications hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated The weighted sum value of the weighted sum value of summing value and stability indicator hundred-mark system score average is weighed, obtains each manufacturer to be evaluated Server comprehensive grading.
Further, in the above method, by each sampling of each single item performance indications of the server of each manufacturer to be evaluated Point is polymerize, and respectively obtains performance index value, including:
Ask for 9,000 of the sampled point in the preset time of each single item performance indications of the server of each manufacturer to be evaluated Quantile, the performance index value as each single item performance indications of the server of each manufacturer to be evaluated.
Further, in the above method, the calculation of 90 percentiles is as follows:
With descending order, each the n in the preset time of each single item performance indications of the server of manufacturer to be evaluated is individual Sampled point, the sequence number of sequence post-sampling point are incremented to n by 1, and n is positive integer;
Gauge index i=np%, wherein, p=90,
If it is not integer that index i, which is calculated, it will adjoin and be more than the sampled point corresponding to i sequence number as described 9,000 Quantile;
If it is integer that index i, which is calculated, by the sampling corresponding to the sampled point corresponding to serial number i and serial number i+l The average value of point sum is as 90 percentile.
Further, in the above method, by each sampling of each single item performance indications of the server of each manufacturer to be evaluated Point is polymerize, and obtains stability indicator value, including:
Using the coefficient of variation as stability indicator, to each single item performance indications of the server of each manufacturer to be evaluated The coefficient of variation of each sampled point in preset time is calculated according to below equation,
Coefficient of variation C.V=(SD ÷ MN) × 100%, wherein, averageStandard deviationxjFor some sampled point, n represents the number of sampled point, and j is one in n sampled point The numbering of sampled point.
Further, in the above method, the evaluation tool include fio, MBW, calculate prime number, calculate pi, One kind or any combination in Ubench.
Further, in the above method, the performance indications of the fio evaluation tools are big including disk operating species/file Any combination of small species/evaluation metricses species.
According to the another aspect of the application, a kind of server evaluation and test equipment is additionally provided, the equipment includes:
Collection device, each single item performance indications of each evaluation tool of the server for collecting each manufacturer to be evaluated Sampled point, wherein, the sampled point of collection is the sampled point in preset time;
Polyplant, for each sampled point of each single item performance indications of the server of each manufacturer to be evaluated to be gathered Close, respectively obtain corresponding performance index value and stability indicator value;
Comprehensive grading device, the performance indications for all performance indications of the server according to each manufacturer to be evaluated Value and stability indicator value, obtain the comprehensive grading of the server of each manufacturer to be evaluated.
Further, in the said equipment, the comprehensive grading device, for the every of the server to each manufacturer to be evaluated The performance index value and stability indicator value of one performance indications are normalized, obtain corresponding to performance indications normalization score Score is normalized with stability indicator;And the performance indications of all performance indications according to the server of each manufacturer to be evaluated Score and stability indicator normalization score are normalized, obtains the comprehensive grading of the server of each manufacturer to be evaluated.
Further, in the said equipment, the comprehensive grading device, for all to be evaluated of each single item performance indications The performance index value of the server of manufacturer, with the performance indications of the server of highest manufacturer to be evaluated in all properties desired value It is worth for 100 points, the performance index value of the server of all manufacturers to be evaluated of this performance indications is normalized, obtains every The performance indications hundred-mark system score of each manufacturer to be evaluated of one performance indications;And to all to be evaluated of each single item performance indications Survey manufacturer server stability indicator value, with all stability indicator values the server of highest manufacturer to be evaluated it is steady Qualitative index value is 100 points, and the stability indicator value of the server of all manufacturers to be evaluated of this performance indications is returned One changes, and obtains the stability indicator hundred-mark system score of each manufacturer to be evaluated of each single item performance indications.
Further, in the said equipment, the comprehensive grading device, for calculating the server of each manufacturer to be evaluated The average value of all performance indications hundred-mark system scores of each evaluation tool and putting down for all stability indicator hundred-mark system scores Average, performance indications hundred-mark system score average respectively as each evaluation tool of the server of the manufacturer to be evaluated and stably Property index hundred-mark system score average;And the performance indications percentage of each evaluation tool according to the server of each manufacturer to be evaluated It is made and divides average and stability indicator hundred-mark system score average, obtains the comprehensive grading of the server of each manufacturer to be evaluated.
Further, in the said equipment, the comprehensive grading device, for the server according to each manufacturer to be evaluated The performance indications hundred-mark system score average and stability indicator hundred-mark system score average of all evaluation tools, are obtained each to be evaluated The weighted sum value and stability indicator hundred of the performance indications hundred-mark system score average of all evaluation tools of the server of manufacturer Divide the weighted sum value for being made and dividing average;And the performance indications of all evaluation tools according to the server of each manufacturer to be evaluated The weighted sum value of the weighted sum value of hundred-mark system score average and the weighted sum value of stability indicator hundred-mark system score average, Obtain the comprehensive grading of the server of each manufacturer to be evaluated.
Further, in the said equipment, the polyplant, for ask for each manufacturer to be evaluated server it is each 90 percentiles of the sampled point in the preset time of item performance indications, each single item as the server of each manufacturer to be evaluated The performance index value of performance indications.
Further, in the said equipment, the polyplant is as follows to the calculation of 90 percentiles:
With descending order, each the n in the preset time of each single item performance indications of the server of manufacturer to be evaluated is individual Sampled point, the sequence number of sequence post-sampling point are incremented to n by 1, and n is positive integer;
Gauge index i=np%, wherein, p=90,
If it is not integer that index i, which is calculated, it will adjoin and be more than the sampled point corresponding to i sequence number as described 9,000 Quantile;
If it is integer that index i, which is calculated, by the sampling corresponding to the sampled point corresponding to serial number i and serial number i+l The average value of point sum is as 90 percentile.
Further, in the said equipment, the polyplant, for using the coefficient of variation as stability indicator, to every The coefficient of variation of each sampled point in the preset time of each single item performance indications of the server of individual manufacturer to be evaluated is according to following Formula is calculated,
Coefficient of variation C.V=(SD ÷ MN) × 100%, wherein, averageStandard deviationxjFor some sampled point, n represents the number of sampled point, and j is one in n sampled point The numbering of sampled point.
Further, in the said equipment, the evaluation tool include fio, MBW, calculate prime number, calculate pi, One kind or any combination in Ubench.
Further, in the said equipment, the performance indications of the fio evaluation tools are big including disk operating species/file Any combination of small species/evaluation metricses species.
According to the one side of the application, a kind of server evaluation and test equipment is additionally provided, wherein, the equipment includes:
Processor;
And it is arranged to store the memory of computer executable instructions, the executable instruction makes institute when executed State processor:
The sampled point of the server of manufacturer to be evaluated is collected, wherein, the sampled point of collection is adopting in preset time Sampling point, each sampled point correspond to a performance indications, and each performance indications corresponds to an evaluation tool;
Sampled point is polymerize, the performance index value and stability for respectively obtaining the server of each manufacturer to be evaluated refer to Scale value, wherein, each performance index value corresponds to a performance indications, and each stability indicator value refers to corresponding to a performance Mark;
According to the performance index value and stability indicator value, the comprehensive grading of the server of manufacturer to be evaluated is obtained.
Compared with prior art, the application passes through the every of each evaluation tool for the server for collecting each manufacturer to be evaluated The sampled point of one performance indications, wherein, the sampled point of collection is the sampled point in preset time, will each factory to be evaluated Each sampled point of each single item performance indications of the server of business is polymerize, and respectively obtains corresponding performance index value and stability Desired value, according to the performance index value and stability indicator value of all performance indications of the server of each manufacturer to be evaluated, It can obtain scoring comprising the whole synthesis for evaluating and testing obtained multinomial performance index and stability by multiple evaluation tools, realize weighing apparatus Not only consider performance indications when measuring server performance, while consider stability indicator, so as to the clothes to each manufacturer to be evaluated The performance of business device more comprehensively, accurately evaluate and test.
Brief description of the drawings
By reading the detailed description made to non-limiting example made with reference to the following drawings, the application's is other Feature, objects and advantages will become more apparent upon:
Fig. 1 shows the flow chart of the server evaluating method according to the application one side;
Fig. 2 shows the schematic diagram of the evaluation tool of the preferred embodiment of the application one;
Fig. 3 shows the flow chart of the server evaluating method of the preferred embodiment of the application one;
Fig. 4 shows the flow chart of the server evaluating method of another preferred embodiment of the application;
Fig. 5 shows the flow chart of the server evaluating method of the another preferred embodiment of the application;
Fig. 6 shows the flow chart of the server evaluating method of the another preferred embodiment of the application;
Fig. 7 shows the flow chart of the server evaluating method of one specific application example of the application
Fig. 8 shows to evaluate and test the module map of equipment according to a kind of server of the application other side.
Same or analogous reference represents same or analogous part in accompanying drawing.
Embodiment
The application is described in further detail below in conjunction with the accompanying drawings.
In one typical configuration of the application, terminal, the equipment of service network and trusted party include one or more Processor (CPU), input/output interface, network interface and internal memory.
Internal memory may include computer-readable medium in volatile memory, random access memory (RAM) and/or The forms such as Nonvolatile memory, such as read-only storage (ROM) or flash memory (flash RAM).Internal memory is computer-readable medium Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method Or technology come realize information store.Information can be computer-readable instruction, data structure, the module of program or other data. The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), moved State random access memory (DRAM), other kinds of random access memory (RAM), read-only storage (ROM), electric erasable Programmable read only memory (EEPROM), fast flash memory bank or other memory techniques, read-only optical disc read-only storage (CD-ROM), Digital versatile disc (DVD) or other optical storages, magnetic cassette tape, magnetic disk storage or other magnetic storage apparatus or Any other non-transmission medium, the information that can be accessed by a computing device available for storage.Defined according to herein, computer Computer-readable recording medium does not include non-temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
As shown in figure 1, the application provides a kind of server evaluating method, this method includes:
Step S11, each single item performance indications of each evaluation tool of the server of each manufacturer to be evaluated of collection are adopted Sampling point, wherein, the sampled point of collection is the sampled point in preset time;Here, in the server of each manufacturer to be evaluated All parts such as CPU, disk and internal memory etc. can be tested and assessed respectively by multiple evaluation tools, each evaluation tool can be set again One or more performance indications are put, can be with for example, in order to respectively obtain performance index value and stability indicator value in subsequent step The preset time of each single item performance indications of each evaluation tool of the server of each manufacturer to be evaluated is collected as in 1 day Sampled point, the initial data of performance index value is obtained as follow-up polymerization, furthermore it is possible to collect each manufacturer to be evaluated The preset time of each single item performance indications of each evaluation tool of server is the sampled point in 7 days, as to be follow-up Polymerization obtains the initial data of stability indicator value;
Step S12, each sampled point of each single item performance indications of the server of each manufacturer to be evaluated is polymerize, point Performance index value and stability indicator value corresponding to not obtaining;Here, it is by each single item of the server of each manufacturer to be evaluated Each sampled point is polymerize in the preset time of performance indications, respectively obtains each single item of the server of each manufacturer to be evaluated A performance index value and stability indicator value in the preset time of energy index;
Step S13, according to the performance index value of all performance indications of the server of each manufacturer to be evaluated and stably Property desired value, obtains the comprehensive grading of the server of each manufacturer to be evaluated.Here, the server according to each manufacturer to be evaluated All evaluation tools all performance indications performance index value and stability indicator value, can obtain include by multiple surveys The whole synthesis scoring of the tool obtained multinomial performance index of evaluation and test of evaluating somebody's work and stability, realize when weighing server performance not only Consider performance indications, while consider stability indicator, so as to the performance of the server to each manufacturer to be evaluated carry out more comprehensively, Accurately evaluate and test.
As shown in Fig. 2 in the preferred embodiment of server evaluating method one of the application, the test and appraisal in step S11 Instrument includes fio, MBW, calculates prime number, calculates pi, one kind or any combination in Ubench.Specifically, calculate prime number Instrument such as sysbench, sysbench be one increase income, be modular, cross-platform multi-threading performance testing tool, can With for carrying out the performance test of CPU, internal memory, magnetic disc i/o, thread, database;
Fio is the instrument for testing disk, for carrying out pressure test and checking to hardware, supports 13 kinds of different I/O to draw Holding up can be by different units block (blocksize) size, different read-write mode (random, sequential), different team Row depth (iodepth), different engine (engine:Either synchronously or asynchronously) come test data disk I/O abilities and handling capacity;
Ubench is a evaluation and test machine CPU and the instrument of internal memory (Memory) performance, is commented for CPU Ubench performances It is the computational methods mixed using integer and floating number to divide, and the amount of calculation of floating number and the amount of calculation of integer ratio are 1:3, CPU Ubench Performance Scores it is higher, cpu performance is better, and equally, the Ubench Performance Scores of internal memory are higher, and Memory performances are got over It is good;
MBW be it is a can be with the instrument of test memory replication bandwidth, it can obtain tri- kinds of MEMCPY, DUMP, MCBLCOK The memory bandwidth of mode, MBW is assigned with two memory arrays in test process, and then data are copied to from a memory array Another, obtained memory bandwidth value is exactly the lower memory data copy speed that can reach of three kinds of modes of operation.
Detailed, as shown in Fig. 2 in step S11, server such as cloud service of the CPU evaluating tools to each manufacturer can be used The CPU of device is evaluated and tested, and the sampled point of each single item performance indications in persistent collection preset time such as 1 day and 7 days enters number According to storehouse, CPU evaluation tool includes calculating prime number, calculates tri- instruments of pi and Ubench, wherein,
The instrument for calculating pi is used for each of a performance indications for collecting the single thread 5000 circle of position frequency times of calculating Sampled point;
The instrument for calculating prime number is calculated in 10000 for collecting prime number time, 2 threads in single thread calculating 10000 respectively Prime number time and 4 threads calculate each sampled point of three performance indications of prime number time in 10000;
Ubench instruments are used for each sampled point for collecting a performance indications of single thread Ubench scorings;
In addition, as shown in Fig. 2 disk of the fio disks evaluating tool to the server such as Cloud Server of each manufacturer can be used Evaluated and tested, and the sampled point of each single item performance indications in persistent collection preset time such as 1 day and 7 days enters database, often One performance indications can be any combination of disk operating species/file size species/evaluation metricses species, wherein, disk Operation species include random write and the class of random write two, and respective file size species may include 1K, 4K, 16K, 64K, 256K, 1M six Class, evaluation metricses species include IOPS (number per second for being written and read (I/O) operation), delay, the class of bandwidth three;
In addition, as shown in Fig. 2 internal memory evaluating tool can be used to carry out the internal memory of the server such as Cloud Server of each manufacturer Evaluation and test, and collect index and carry out data loading, and each single item in persistent collection preset time such as 1 day and 7 days as sampled point The sampled point of performance indications enters database, and internal memory evaluating tool includes MBW and Ubench binomials, wherein, the performance of MBW instruments Index includes MEMCPY memory bandwidths, DUMB memory bandwidth MCBLOCK memory bandwidths three, the performance indications bag of Ubench instruments The scoring of Ubench internal memories is included, here, each sampled point storage data field format of each performance indications is as follows:
Manufacturer:Method of testing:Index name:Desired value:Index unit:Entry time, such as such as aliyun:MBW: memcpy:100:MB/S:1449936598。
As shown in figure 3, in the preferred embodiment of server evaluating method one of the application, step S13, according to each to be evaluated The performance index value and stability indicator value of all performance indications of the server of manufacturer are surveyed, obtains each manufacturer to be evaluated The comprehensive grading of server, including:
Step S131, to the performance index value and stability of each single item performance indications of the server of each manufacturer to be evaluated Desired value is normalized, and obtains corresponding performance indications normalization score and stability indicator normalization score;
Step S132, normalized according to the performance indications of all performance indications of the server of each manufacturer to be evaluated Divide and stability indicator normalizes score, obtain the comprehensive grading of the server of each manufacturer to be evaluated.Here, by by each The performance index value and stability indicator value of each single item performance indications of the server of manufacturer to be evaluated are normalized, and obtain pair Answer performance indications normalization score and stability indicator normalization score, be easy to subsequently according to performance indications normalize score and Stability indicator normalization score obtains the comprehensive grading of the server of the manufacturer to be evaluated of unified criterion, and is easy to final The comprehensive grading of the server of different manufacturers to be evaluated is intuitively compared.
As shown in figure 4, in the preferred embodiment of server evaluating method one of the application, step S131, to each to be evaluated The performance index value and stability indicator value for surveying each single item performance indications of the server of manufacturer are normalized, and obtain corresponding Performance indications normalize score and stability indicator normalization score, including:
Step S1311, to the performance index value of the server of all manufacturers to be evaluated of each single item performance indications, with all The performance index value of the server of highest manufacturer to be evaluated is 100 points in performance index value, is owned to this performance indications The performance index value of the server of manufacturer to be evaluated is normalized, and obtains each manufacturer to be evaluated of each single item performance indications Performance indications hundred-mark system score;
Step S1312, to the stability indicator value of the server of all manufacturers to be evaluated of each single item performance indications, with institute The stability indicator value for having the server of highest manufacturer to be evaluated in stability indicator value is 100 points, to this performance indications The stability indicator values of server of all manufacturers to be evaluated be normalized, obtain each to be evaluated of each single item performance indications Survey the stability indicator hundred-mark system score of manufacturer.Here, obtained performance indications normalization is scored at performance indications hundred-mark system and obtained Point, corresponding, obtained stability indicator normalization is scored at stability indicator hundred-mark system score, is easy to subsequently according to performance be referred to Mark hundred-mark system score and stability indicator hundred-mark system score obtain the synthesis of the server of the manufacturer to be evaluated of unified criterion Scoring, and it is easy to most that the comprehensive grading of the server of different manufacturers to be evaluated is more intuitively compared at last.Specifically, one Evaluation tool is used to test and assess one or more performance indications, evaluation tool may include fio, MBW, calculate prime number, calculate pi, Ubench etc..
As shown in figure 5, in the preferred embodiment of server evaluating method one of the application, step S132, treated according to each The performance indications normalization score and stability indicator normalization score of all performance indications of the server of manufacturer are evaluated and tested, is obtained To the comprehensive grading of the server of each manufacturer to be evaluated, including:
Step S1321, calculate all performance indications hundred of each evaluation tool of the server of each manufacturer to be evaluated Divide the average value that the average value divided and all stability indicator hundred-mark system scores is made, respectively as the service of the manufacturer to be evaluated The performance indications hundred-mark system score average and stability indicator hundred-mark system score average of each evaluation tool of device;
Step S1322, obtained according to the performance indications hundred-mark system of each evaluation tool of the server of each manufacturer to be evaluated Divide average and stability indicator hundred-mark system score average, the comprehensive grading of the server of each manufacturer to be evaluated is obtained, so as to protect Demonstrate,prove obtained comprehensive grading more comprehensively, accurately.
As shown in fig. 6, in the preferred embodiment of server evaluating method one of the application, step S1322, treated according to each Evaluate and test the performance indications hundred-mark system score average and stability indicator hundred-mark system score of each evaluation tool of the server of manufacturer Average, the comprehensive grading of the server of the manufacturer to be evaluated is obtained, including:
Step S13221, obtained according to the performance indications hundred-mark system of all evaluation tools of the server of each manufacturer to be evaluated Divide average, obtain the weighting of the performance indications hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated Summing value, according to the stability indicator hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated, obtain To the weighted sum value of the stability indicator hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated; Specifically, the weighted sum of the performance indications hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated Value can be calculated according to following manner, and weight 0.2 is assigned to each evaluation tool, calculate the institute of the server of each manufacturer to be evaluated There is a weighted sum value of the performance indications hundred-mark system score average of evaluation tool, formula is as follows:
The weighting of the performance indications hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated is asked With the performance indications hundred-mark system score average of the performance indications hundred-mark system score average+0.2*MBW instruments of value=0.2*fio instruments + 0.2* calculates the performance indications hundred-mark system of the performance indications hundred-mark system score average+0.2* calculating pi instruments of prime number instrument The performance indications hundred-mark system score average of score average+0.2*Ubench instruments;
In addition, the stability indicator hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated Weighted sum value can be calculated according to following manner, and weight 0.2 is assigned to each evaluation tool, calculate the clothes of each manufacturer to be evaluated The weighted sum value of the stability indicator hundred-mark system score average of all evaluation tools of business device, formula are as follows:
The weighting of the stability indicator hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated The stability indicator hundred-mark system of the stability indicator hundred-mark system score average+0.2*MBW instruments of summing value=0.2*fio instruments obtains The stability indicator hundred-mark system score average+0.2* for dividing average+0.2* to calculate prime number instrument calculates the stability of pi instrument The stability indicator hundred-mark system score average of index hundred-mark system score average+0.2*Ubench instruments;
Step S13222, obtained according to the performance indications hundred-mark system of all evaluation tools of the server of each manufacturer to be evaluated Divide the weighted sum value of the weighted sum value of average and the weighted sum value of stability indicator hundred-mark system score average, obtain each The comprehensive grading of the server of manufacturer to be evaluated, so as to ensure obtained comprehensive grading more comprehensively, accurately.It is here, each to be evaluated The calculation formula for surveying the comprehensive grading of the server of manufacturer is as follows:
All evaluation tools of the server of comprehensive grading=0.5* manufacturers to be evaluated of the server of manufacturer to be evaluated The stabilization of all evaluation tools of the server of the weighted sum value+0.5* manufacturers to be evaluated of performance indications hundred-mark system score average The weighted sum value of property index hundred-mark system score average.
In the preferred embodiment of server evaluating method one of the application, in step S12, by each manufacturer to be evaluated Each sampled point of each single item performance indications of server is polymerize, and respectively obtains performance index value, including:
Ask for 9,000 of the sampled point in the preset time of each single item performance indications of the server of each manufacturer to be evaluated Quantile, as the performance index value of each single item performance indications of the server of each manufacturer to be evaluated, obtained so as to accurate The performance index value of each single item performance indications after to polymerization.Here, the preset time can be in 1 day.
In the preferred embodiment of server evaluating method one of the application, the calculation of 90 percentiles is as follows:
Arrange each single item performance indications of the server of each manufacturer to be evaluated from small to large with descending order N sampled point in preset time, the sequence number of sequence post-sampling point are incremented to n by 1, and n is positive integer;
Gauge index i=np%, wherein, p=90,
If it is not integer that index i, which is calculated, it will adjoin and be more than the sampled point corresponding to i sequence number as described 9,000 Quantile, for example, n=9 sampled point is not integer with such as table 1 of the result after descending order, i=8.1, then will adjoin and Sampled point 90 corresponding to sequence number 9 more than 8.1 is used as 90 percentile;
Sequence number 1 2 3 4 5 6 7 8 9
Sampled point 10 20 30 40 50 60 70 80 90
Table 1
If it is integer that index i, which is calculated, by the sampling corresponding to the sampled point corresponding to serial number i and serial number i+l The average value of point sum is as 90 percentile, for example, n=10 sampled point is with such as table of the result after descending order 2, i=9 be integer, then by the average value of the sampled point sum 190 corresponding to the sampled point corresponding to serial number 9 and serial number 10 95 are used as 90 percentile.
Sequence number 1 2 3 4 5 6 7 8 9 10
Sampled point 10 20 30 40 50 60 70 80 90 100
Table 2
In the present embodiment, percentile calculation divides following three step:
A. with descending order initial data.
B. gauge index i=np%, p=90
C. value
If a) i is not integer, i is rounded up, adjoins the position that integer is pth percentile more than i.
If b) i is integer, pth percentile is i-th average value with (i+l) item data.
In the preferred embodiment of server evaluating method one of the application, in step S12, by each manufacturer to be evaluated Each sampled point of each single item performance indications of server is polymerize, and obtains stability indicator value, including:
Using the coefficient of variation as stability indicator, to each single item performance indications of the server of each manufacturer to be evaluated The coefficient of variation of each sampled point in preset time is calculated according to below equation,
Coefficient of variation C.V=(SD ÷ MN) × 100%, wherein, averageStandard deviationxjFor some sampled point, n represents the number of sampled point, and j is one in n sampled point The numbering of sampled point.The present embodiment can be by respectively adopting in 7 days of each single item performance indications of the server of each manufacturer to be evaluated Sampling point is polymerize.
As shown in fig. 7, in one specific application example of the application, server evaluating method comprises the following steps:
Step S1, collect the sampling of each single item performance indications of each evaluation tool of the server of each manufacturer to be evaluated Point, wherein, the sampled point of collection is the sampled point in preset time;
Step S2, each sampled point in 1 day of each single item performance indications of the server of each manufacturer to be evaluated is gathered Close, obtain 90 percentiles as performance index value;
Step S3, each sampled point in 7 days of each single item performance indications of the server of each manufacturer to be evaluated is gathered Close, obtain C.V values as stability indicator value;
Step S4, it is sexual with institute to the performance index value of the server of all manufacturers to be evaluated of each single item performance indications The performance index value of the server of highest manufacturer to be evaluated is 100 points in energy desired value, to being needed for this performance indications The performance index value for evaluating and testing the server of manufacturer is normalized, and obtains the property of each manufacturer to be evaluated of each single item performance indications Can index hundred-mark system score;To the stability indicator value of the server of all manufacturers to be evaluated of each single item performance indications, with institute The stability indicator value for having the server of highest manufacturer to be evaluated in stability indicator value is 100 points, to this performance indications The stability indicator values of server of all manufacturers to be evaluated be normalized, obtain each to be evaluated of each single item performance indications Survey the stability indicator hundred-mark system score of manufacturer;Calculate all items of each evaluation tool of the server of each manufacturer to be evaluated The average value of the average value of performance indications hundred-mark system score and all stability indicator hundred-mark system scores, it is to be evaluated respectively as this The performance indications hundred-mark system score average and stability indicator hundred-mark system score average of each evaluation tool of the server of manufacturer;
Step S5, it is equal according to the performance indications hundred-mark system score of all evaluation tools of the server of each manufacturer to be evaluated Value, obtains the weighted sum of the performance indications hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated Value;
Step S6, according to the stability indicator hundred-mark system score of all evaluation tools of the server of each manufacturer to be evaluated Average, obtain the weighting of the stability indicator hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated Summing value;
Step S7, it is equal according to the performance indications hundred-mark system score of all evaluation tools of the server of each manufacturer to be evaluated The weighted sum value of the weighted sum value of value and the weighted sum value of stability indicator hundred-mark system score average, is obtained each to be evaluated Survey the comprehensive grading of the server of manufacturer.
As shown in figure 8, also providing a kind of server according to the another side of the application evaluates and tests equipment, the equipment 100 includes:
Collection device 1, each single item performance of each evaluation tool of the server for collecting each manufacturer to be evaluated refer to Target sampled point, wherein, the sampled point of collection is the sampled point in preset time;Here, the clothes of each manufacturer to be evaluated All parts such as CPU, disk and internal memory etc. in business device can be tested and assessed by multiple evaluation tools respectively, each evaluation tool One or more performance indications can be set again, for example, in order to respectively obtain performance index value and stability indicator in subsequent step Value, the preset time of each single item performance indications of each evaluation tool of the server of each manufacturer to be evaluated can be collected For the sampled point in 1 day, the initial data of performance index value is obtained as follow-up polymerization, furthermore it is possible to collect each to be evaluated The preset time for surveying each single item performance indications of each evaluation tool of the server of manufacturer is the sampled point in 7 days, is made To obtain the initial data of stability indicator value for follow-up polymerization;
Polyplant 2, for each sampled point of each single item performance indications of the server of each manufacturer to be evaluated to be carried out Polymerization, respectively obtain corresponding performance index value and stability indicator value;Here, it is by the server of each manufacturer to be evaluated Each sampled point is polymerize in the preset time of each single item performance indications, respectively obtain each manufacturer to be evaluated server it is every A performance index value and stability indicator value in the preset time of one performance indications;
Comprehensive grading device 3, the performance for all performance indications of the server according to each manufacturer to be evaluated refer to Scale value and stability indicator value, obtain the comprehensive grading of the server of each manufacturer to be evaluated.Here, according to each factory to be evaluated The performance index value and stability indicator value of all performance indications of all evaluation tools of the server of business, can be wrapped Scored containing the whole synthesis that obtained multinomial performance index and stability is evaluated and tested by multiple evaluation tools, realize and weigh server Not only consider performance indications during energy, while consider stability indicator, so as to the performance of the server to each manufacturer to be evaluated Progress more comprehensively, is accurately evaluated and tested.
As shown in Fig. 2 in a kind of server evaluation and test preferred embodiment of equipment one that the application provides, the evaluation tool Including one kind or any combination in fio, MBW, calculating prime number, calculating pi, Ubench.Specifically, calculate the work of prime number Tool such as sysbench, sysbench be one increase income, be modular, cross-platform multi-threading performance testing tool, Ke Yiyong To carry out the performance test of CPU, internal memory, magnetic disc i/o, thread, database;
Fio is the instrument for testing disk, for carrying out pressure test and checking to hardware, supports 13 kinds of different I/O to draw Holding up can be by different units block (blocksize) size, different read-write mode (random, sequential), different team Row depth (iodepth), different engine (engine:Either synchronously or asynchronously) come test data disk I/O abilities and handling capacity;
Ubench is a evaluation and test machine CPU and the instrument of internal memory (Memory) performance, is commented for CPU Ubench performances It is the computational methods mixed using integer and floating number to divide, and the amount of calculation of floating number and the amount of calculation of integer ratio are 1:3, CPU Ubench Performance Scores it is higher, cpu performance is better, and equally, the Ubench Performance Scores of internal memory are higher, and Memory performances are got over It is good;
MBW be it is a can be with the instrument of test memory replication bandwidth, it can obtain tri- kinds of MEMCPY, DUMP, MCBLCOK The memory bandwidth of mode, MBW is assigned with two memory arrays in test process, and then data are copied to from a memory array Another, obtained memory bandwidth value is exactly the lower memory data copy speed that can reach of three kinds of modes of operation.
Detailed, as shown in Fig. 2 in collection device 1, it CPU evaluating tools can be used to take the server such as cloud of each manufacturer The CPU of business device is evaluated and tested, and the sampled point of each single item performance indications in persistent collection preset time such as 1 day and 7 days enters Database, CPU evaluation tool include calculating prime number, calculate tri- instruments of pi and Ubench, wherein,
The instrument for calculating pi is used for each of a performance indications for collecting the single thread 5000 circle of position frequency times of calculating Sampled point;
The instrument for calculating prime number is calculated in 10000 for collecting prime number time, 2 threads in single thread calculating 10000 respectively Prime number time and 4 threads calculate each sampled point of three performance indications of prime number time in 10000;
Ubench instruments are used for each sampled point for collecting a performance indications of single thread Ubench scorings;
In addition, as shown in Fig. 2 disk of the fio disks evaluating tool to the server such as Cloud Server of each manufacturer can be used Evaluated and tested, and the sampled point of each single item performance indications in persistent collection preset time such as 1 day and 7 days enters database, often One performance indications can be any combination of disk operating species/file size species/evaluation metricses species, wherein, disk Operation species include random write and the class of random write two, and respective file size species may include 1K, 4K, 16K, 64K, 256K, 1M six Class, evaluation metricses species include IOPS (number per second for being written and read (I/O) operation), delay, the class of bandwidth three;
In addition, as shown in Fig. 2 internal memory evaluating tool can be used to carry out the internal memory of the server such as Cloud Server of each manufacturer Evaluation and test, and collect index and carry out data loading, and each single item in persistent collection preset time such as 1 day and 7 days as sampled point The sampled point of performance indications enters database, and internal memory evaluating tool includes MBW and Ubench binomials, wherein, the performance of MBW instruments Index includes MEMCPY memory bandwidths, DUMB memory bandwidth MCBLOCK memory bandwidths three, the performance indications bag of Ubench instruments The scoring of Ubench internal memories is included, here, each sampled point storage data field format of each performance indications is as follows:
Manufacturer:Method of testing:Index name:Desired value:Index unit:Entry time, such as such as aliyun:MBW: memcpy:100:MB/S:1449936598。
In a kind of server evaluation and test preferred embodiment of equipment one that the application provides, the comprehensive grading device 3, it is used for The performance index value and stability indicator value of each single item performance indications of the server of each manufacturer to be evaluated are normalized, Performance indications normalization score corresponding to obtaining and stability indicator normalization score;And the service according to each manufacturer to be evaluated The performance indications normalization score and stability indicator normalization score of all performance indications of device, obtain each factory to be evaluated The comprehensive grading of the server of business.Here, by by the performance of each single item performance indications of the server of each manufacturer to be evaluated Desired value and stability indicator value are normalized, and obtain corresponding performance indications normalization score and stability indicator normalization Score, it is easy to subsequently normalize score according to performance indications and stability indicator normalization score obtains treating for unified criterion The comprehensive grading of the server of manufacturer is evaluated and tested, and is easy to the comprehensive grading of the most server of different manufacturers to be evaluated at last to carry out directly The comparison of sight.
In a kind of server evaluation and test preferred embodiment of equipment one that the application provides, the comprehensive grading device 3, it is used for To the performance index value of the server of all manufacturers to be evaluated of each single item performance indications, with highest in all properties desired value The performance index value of the server of manufacturer to be evaluated is 100 points, to the server of all manufacturers to be evaluated of this performance indications Performance index value be normalized, the performance indications hundred-mark system for obtaining each manufacturer to be evaluated of each single item performance indications obtains Point;And the stability indicator value of the server to all manufacturers to be evaluated of each single item performance indications, with all stability indicators The stability indicator value of the server of highest manufacturer to be evaluated is 100 points in value, to all to be evaluated of this performance indications The stability indicator value of the server of manufacturer is normalized, and obtains the stabilization of each manufacturer to be evaluated of each single item performance indications Property index hundred-mark system score.Here, obtained performance indications normalization is scored at performance indications hundred-mark system score, and it is corresponding, obtain To stability indicator normalization be scored at stability indicator hundred-mark system score, be easy to subsequently according to performance indications hundred-mark system score The comprehensive grading of the server of the manufacturer to be evaluated of unified criterion is obtained with stability indicator hundred-mark system score, and is easy to most The comprehensive grading of the server of different manufacturers to be evaluated is more intuitively compared at last.Specifically, an evaluation tool is used for Test and assess one or more performance indications, evaluation tool may include fio, MBW, calculate prime number, calculate pi, Ubench etc..
In a kind of server evaluation and test preferred embodiment of equipment one that the application provides, the comprehensive grading device 3, it is used for Calculate all performance indications hundred-mark system scores of each evaluation tool of the server of each manufacturer to be evaluated average value and The average value of all stability indicator hundred-mark system scores, respectively as each evaluation tool of the server of the manufacturer to be evaluated Performance indications hundred-mark system score average and stability indicator hundred-mark system score average;And the server according to each manufacturer to be evaluated Each evaluation tool performance indications hundred-mark system score average and stability indicator hundred-mark system score average, obtain each to be evaluated The comprehensive grading of the server of manufacturer is surveyed, so as to ensure obtained comprehensive grading more comprehensively, accurately.
In a kind of server evaluation and test preferred embodiment of equipment one that the application provides, the comprehensive grading device 3, it is used for According to the performance indications hundred-mark system score average and stability indicator of all evaluation tools of the server of each manufacturer to be evaluated Hundred-mark system score average, the performance indications hundred-mark system score for obtaining all evaluation tools of the server of each manufacturer to be evaluated are equal The weighted sum value of value and the weighted sum value of stability indicator hundred-mark system score average;And the clothes according to each manufacturer to be evaluated The weighted sum value and stability indicator hundred-mark system score of the performance indications hundred-mark system score average of all evaluation tools of business device The weighted sum value of the weighted sum value of average, obtain the comprehensive grading of the server of each manufacturer to be evaluated.It is specifically, each The weighted sum value of the performance indications hundred-mark system score average of all evaluation tools of the server of manufacturer to be evaluated can be according to such as Under type is calculated, and weight 0.2 is assigned to each evaluation tool, calculates all evaluation tools of the server of each manufacturer to be evaluated Performance indications hundred-mark system score average weighted sum value, formula is as follows:
The weighting of the performance indications hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated is asked With the performance indications hundred-mark system score average of the performance indications hundred-mark system score average+0.2*MBW instruments of value=0.2*fio instruments + 0.2* calculates the performance indications hundred-mark system of the performance indications hundred-mark system score average+0.2* calculating pi instruments of prime number instrument The performance indications hundred-mark system score average of score average+0.2*Ubench instruments;
In addition, the stability indicator hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated Weighted sum value can be calculated according to following manner, and weight 0.2 is assigned to each evaluation tool, calculate the clothes of each manufacturer to be evaluated The weighted sum value of the stability indicator hundred-mark system score average of all evaluation tools of business device, formula are as follows:
The weighting of the stability indicator hundred-mark system score average of all evaluation tools of the server of each manufacturer to be evaluated The stability indicator hundred-mark system of the stability indicator hundred-mark system score average+0.2*MBW instruments of summing value=0.2*fio instruments obtains The stability indicator hundred-mark system score average+0.2* for dividing average+0.2* to calculate prime number instrument calculates the stability of pi instrument The stability indicator hundred-mark system score average of index hundred-mark system score average+0.2*Ubench instruments;
In addition, the performance indications hundred-mark system score average according to all evaluation tools of the server of each manufacturer to be evaluated Weighted sum value and stability indicator hundred-mark system score average weighted sum value weighted sum value, obtain each to be evaluated The comprehensive grading of the server of manufacturer, so as to ensure obtained comprehensive grading more comprehensively, accurately.Here, each manufacturer to be evaluated Server comprehensive grading calculation formula it is as follows:
All evaluation tools of the server of comprehensive grading=0.5* manufacturers to be evaluated of the server of manufacturer to be evaluated The stabilization of all evaluation tools of the server of the weighted sum value+0.5* manufacturers to be evaluated of performance indications hundred-mark system score average The weighted sum value of property index hundred-mark system score average.
In a kind of server evaluation and test preferred embodiment of equipment one that the application provides, the polyplant 2, for asking for 90 percentiles of the sampled point in the preset time of each single item performance indications of the server of each manufacturer to be evaluated, as every The performance index value of each single item performance indications of the server of individual manufacturer to be evaluated, it is each after polymerization so as to be accurately obtained The performance index value of item performance indications.Here, the preset time can be in 1 day.
In a kind of server evaluation and test preferred embodiment of equipment one that the application provides, the polyplant is to 90 hundredths Several calculations is as follows:
With descending order, each the n in the preset time of each single item performance indications of the server of manufacturer to be evaluated is individual Sampled point, the sequence number of sequence post-sampling point are incremented to n by 1, and n is positive integer;
Gauge index i=np%, wherein, p=90,
If it is not integer that index i, which is calculated, it will adjoin and be more than the sampled point corresponding to i sequence number as described 9,000 Quantile, for example, n=9 sampled point is not integer with such as table 1 of the result after descending order, i=8.1, then will adjoin and Sampled point 90 corresponding to sequence number 9 more than 8.1 is used as 90 percentile;
Table 1
If it is integer that index i, which is calculated, by the sampling corresponding to the sampled point corresponding to serial number i and serial number i+l The average value of point sum is as 90 percentile, for example, n=10 sampled point is with such as table of the result after descending order 2, i=9 be integer, then by the average value of the sampled point sum 190 corresponding to the sampled point corresponding to serial number 9 and serial number 10 95 are used as 90 percentile.
Sequence number 1 2 3 4 5 6 7 8 9 10
Sampled point 10 20 30 40 50 60 70 80 90 100
Table 2
In the present embodiment, percentile calculation divides following three step:
A. with descending order initial data.
B. gauge index i=np%, p=90
C. value
If a) i is not integer, i is rounded up, the integer that adjoins more than i is pth percentage
The position of digit.
If i is integer, pth percentile is i-th average value with (i+l) item data.
In a kind of server evaluation and test preferred embodiment of equipment one that the application provides, the polyplant 2, for using The coefficient of variation is as stability indicator, in the preset time of each single item performance indications of the server of each manufacturer to be evaluated The coefficient of variation of each sampled point is calculated according to below equation,
Coefficient of variation C.V=(SD ÷ MN) × 100%, wherein, averageStandard deviationxjFor some sampled point, n represents the number of sampled point, and j is one in n sampled point The numbering of sampled point.The present embodiment can be by respectively adopting in 7 days of each single item performance indications of the server of each manufacturer to be evaluated Sampling point is polymerize.
In summary, each single item that the application passes through each evaluation tool of the server of each manufacturer to be evaluated of collection The sampled point of energy index, wherein, the sampled point of collection is the sampled point in preset time, by the clothes of each manufacturer to be evaluated Each sampled point of each single item performance indications of business device is polymerize, and respectively obtains corresponding performance index value and stability indicator Value, can be with according to the performance index value and stability indicator value of all performance indications of the server of each manufacturer to be evaluated Obtain scoring comprising the whole synthesis for evaluating and testing obtained multinomial performance index and stability by multiple evaluation tools, realize and weigh clothes Not only consider performance indications during device performance of being engaged in, while consider stability indicator, so as to the server to each manufacturer to be evaluated Performance carry out more comprehensively, accurately evaluate and test.
In addition, present invention also provides a kind of server to evaluate and test equipment, wherein, the equipment includes:
Processor;
And it is arranged to store the memory of computer executable instructions, the executable instruction makes institute when executed State processor:
The sampled point of the server of manufacturer to be evaluated is collected, wherein, the sampled point of collection is adopting in preset time Sampling point, each sampled point correspond to a performance indications, and each performance indications corresponds to an evaluation tool;
Sampled point is polymerize, the performance index value and stability for respectively obtaining the server of each manufacturer to be evaluated refer to Scale value, wherein, each performance index value corresponds to a performance indications, and each stability indicator value refers to corresponding to a performance Mark;
According to the performance index value and stability indicator value, the comprehensive grading of the server of manufacturer to be evaluated is obtained.
Obviously, those skilled in the art can carry out the essence of various changes and modification without departing from the application to the application God and scope.So, if these modifications and variations of the application belong to the scope of the application claim and its equivalent technologies Within, then the application is also intended to comprising including these changes and modification.
It should be noted that the application can be carried out in the assembly of software and/or software and hardware, for example, can adopt With application specific integrated circuit (ASIC), general purpose computer or any other realized similar to hardware device.In one embodiment In, the software program of the application can realize steps described above or function by computing device.Similarly, the application Software program (including related data structure) can be stored in computer readable recording medium storing program for performing, for example, RAM memory, Magnetically or optically driver or floppy disc and similar devices.In addition, some steps or function of the application can employ hardware to realize, example Such as, coordinate as with processor so as to perform the circuit of each step or function.
In addition, the part of the application can be applied to computer program product, such as computer program instructions, when its quilt When computer performs, by the operation of the computer, it can call or provide according to the present processes and/or technical scheme. And the programmed instruction of the present processes is called, it is possibly stored in fixed or moveable recording medium, and/or pass through Broadcast or the data flow in other signal bearing medias and be transmitted, and/or be stored according to described program instruction operation In the working storage of computer equipment.Here, including a device according to one embodiment of the application, the device includes using Memory in storage computer program instructions and processor for execute program instructions, wherein, when the computer program refers to When order is by the computing device, method and/or skill of the plant running based on foregoing multiple embodiments according to the application are triggered Art scheme.
It is obvious to a person skilled in the art that the application is not limited to the details of above-mentioned one exemplary embodiment, Er Qie In the case of without departing substantially from spirit herein or essential characteristic, the application can be realized in other specific forms.Therefore, no matter From the point of view of which point, embodiment all should be regarded as exemplary, and be nonrestrictive, scope of the present application is by appended power Profit requires rather than described above limits, it is intended that all in the implication and scope of the equivalency of claim by falling Change is included in the application.Any reference in claim should not be considered as to the involved claim of limitation.This Outside, it is clear that the word of " comprising " one is not excluded for other units or step, and odd number is not excluded for plural number.That is stated in device claim is multiple Unit or device can also be realized by a unit or device by software or hardware.The first, the second grade word is used for table Show title, and be not offered as any specific order.

Claims (21)

1. a kind of server evaluating method, wherein, this method includes:
The sampled point of the server of manufacturer to be evaluated is collected, wherein, the sampled point of collection is the sampled point in preset time, Each sampled point corresponds to a performance indications, and each performance indications corresponds to an evaluation tool;
Sampled point is polymerize, obtains the performance index value and stability indicator value of the server of manufacturer to be evaluated, wherein, often Individual performance index value corresponds to a performance indications, and each stability indicator value corresponds to a performance indications;
According to the performance index value and stability indicator value, the comprehensive grading of the server of manufacturer to be evaluated is obtained.
2. according to the method for claim 1, wherein, according to the performance index value and stability indicator value, obtain to be evaluated The comprehensive grading of the server of manufacturer is surveyed, including:
The performance index value and stability indicator value are normalized, obtain the performance indications of the server of manufacturer to be evaluated Normalize score and stability indicator normalization score;
Score and stability indicator normalization score are normalized according to the performance indications, obtains the server of manufacturer to be evaluated Comprehensive grading.
3. according to the method for claim 2, wherein, the performance index value and stability indicator value are normalized, The performance indications normalization score and stability indicator normalization score of the server of manufacturer to be evaluated are obtained, including:
Using the performance index value of the server of highest manufacturer to be evaluated as 100 points, to the performance of the server of manufacturer to be evaluated Desired value is normalized, and obtains the performance indications hundred-mark system score of manufacturer to be evaluated;
Using the stability indicator value of the server of highest manufacturer to be evaluated as 100 points, to the steady of the server of manufacturer to be evaluated Qualitative index value is normalized, and obtains the stability indicator hundred-mark system score of manufacturer to be evaluated.
4. according to the method for claim 3, wherein, score and stability indicator normalizing are normalized according to the performance indications Change score, obtain the comprehensive grading of the server of manufacturer to be evaluated, including:
Calculate the average value and stability indicator hundred-mark system score of the performance indications hundred-mark system score of the server of manufacturer to be evaluated Average value, obtained respectively as the server performance index hundred-mark system score average and stability indicator hundred-mark system of manufacturer to be evaluated Divide average;
According to the performance indications hundred-mark system score average of the server of manufacturer to be evaluated and stability indicator hundred-mark system score average, Obtain the comprehensive grading of the server of each manufacturer to be evaluated.
5. the method according to claim 11, wherein, according to the performance indications hundred-mark system score of the server of manufacturer to be evaluated Average and stability indicator hundred-mark system score average, obtain the comprehensive grading of the server of each manufacturer to be evaluated, including:
According to the performance indications hundred-mark system score average of the server of manufacturer to be evaluated and stability indicator hundred-mark system score average, Obtain the weighted sum value and stability indicator hundred-mark system of the performance indications hundred-mark system score average of the server of manufacturer to be evaluated The weighted sum value of score average;
According to the weighted sum value and stability indicator hundred of the performance indications hundred-mark system score average of the server of manufacturer to be evaluated Divide the weighted sum value that the weighted sum value for dividing average is made, obtain the comprehensive grading of the server of manufacturer to be evaluated.
6. according to the method for claim 1, wherein, sampled point is polymerize, obtains the server of manufacturer to be evaluated Performance index value, including:
90 percentiles of the sampled point of the server of manufacturer to be evaluated are asked for, the performance as the server of manufacturer to be evaluated refers to Scale value.
7. according to the method for claim 6, wherein, the calculation of 90 percentiles is as follows:
With descending order each n sampling in the preset time of each single item performance indications of the server of manufacturer to be evaluated Point, the sequence number of sequence post-sampling point are incremented to n by 1, and n is positive integer;
Gauge index i=np%, wherein, p=90,
If it is not integer that index i, which is calculated, it will adjoin and be more than the sampled point corresponding to i sequence number as 90 hundredths Number;
If it is integer that index i, which is calculated, by the sampled point corresponding to serial number i and the sampled point corresponding to serial number i+l it The average value of sum is as 90 percentile.
8. according to the method for claim 1, wherein, sampled point is polymerize, obtains the server of manufacturer to be evaluated Stability indicator value, including:
Using the coefficient of variation as stability indicator, to the coefficient of variation of the sampled point of the server of manufacturer to be evaluated according to following Formula is calculated,
Coefficient of variation C.V=(SD ÷ MN) × 100%, wherein, averageStandard deviationxjFor some sampled point, n represents the number of sampled point, and j is one in n sampled point The numbering of sampled point.
9. according to the method for claim 1, wherein, the evaluation tool includes fio, MBW, calculates prime number, calculates circumference One kind or any combination in rate, Ubench.
10. according to the method for claim 9, wherein, the performance indications of the fio evaluation tools include disk operating kind Any combination of class/file size species/evaluation metricses species.
11. a kind of server evaluates and tests equipment, wherein, the equipment includes:
Collection device, the sampled point of the server for collecting manufacturer to be evaluated, wherein, when the sampled point of collection is presets Interior sampled point, each sampled point correspond to a performance indications, and each performance indications corresponds to an evaluation tool;
Polyplant, for sampled point to be polymerize, respectively obtain the performance index value of the server of each manufacturer to be evaluated With stability indicator value, wherein, each performance index value corresponds to a performance indications, and each stability indicator value corresponds to one Item performance indications;
Comprehensive grading device, for according to the performance index value and stability indicator value, obtaining the server of manufacturer to be evaluated Comprehensive grading.
12. equipment according to claim 11, wherein, the comprehensive grading device, for the performance index value and Stability indicator value is normalized, and obtains the performance indications normalization score and stability indicator of the server of manufacturer to be evaluated Normalize score;And score and stability indicator normalization score are normalized according to the performance indications, obtain manufacturer to be evaluated Server comprehensive grading.
13. equipment according to claim 12, wherein, the comprehensive grading device, for highest manufacturer to be evaluated Server performance index value be 100 points, the performance index value of the server of manufacturer to be evaluated is normalized, treated Evaluate and test the performance indications hundred-mark system score of manufacturer;And using the stability indicator value of the server of highest manufacturer to be evaluated as 100 Point, the stability indicator value of the server of manufacturer to be evaluated is normalized, obtains the stability indicator hundred of manufacturer to be evaluated Divide to be made and divide.
14. equipment according to claim 13, wherein, the comprehensive grading device, for calculating the clothes of manufacturer to be evaluated The average value of performance indications hundred-mark system score and the average value of stability indicator hundred-mark system score of business device, respectively as to be evaluated The server performance index hundred-mark system score average and stability indicator hundred-mark system score average of manufacturer;And according to manufacturer to be evaluated Server performance indications hundred-mark system score average and stability indicator hundred-mark system score average, obtain each manufacturer to be evaluated Server comprehensive grading.
15. equipment according to claim 14, wherein, the comprehensive grading device, for the clothes according to manufacturer to be evaluated The performance indications hundred-mark system score average and stability indicator hundred-mark system score average of business device, obtain the server of manufacturer to be evaluated The weighted sum value of performance indications hundred-mark system score average and the weighted sum value of stability indicator hundred-mark system score average;And According to the weighted sum value and stability indicator hundred-mark system of the performance indications hundred-mark system score average of the server of manufacturer to be evaluated The weighted sum value of the weighted sum value of score average, obtains the comprehensive grading of the server of manufacturer to be evaluated.
16. equipment according to claim 11, wherein, the polyplant, for asking for the server of manufacturer to be evaluated Sampled point 90 percentiles, the performance index value as the server of manufacturer to be evaluated.
17. equipment according to claim 16, wherein, the polyplant is as follows to the calculation of 90 percentiles:
With descending order each n sampling in the preset time of each single item performance indications of the server of manufacturer to be evaluated Point, the sequence number of sequence post-sampling point are incremented to n by 1, and n is positive integer;
Gauge index i=np%, wherein, p=90,
If it is not integer that index i, which is calculated, it will adjoin and be more than the sampled point corresponding to i sequence number as 90 hundredths Number;
If it is integer that index i, which is calculated, by the sampled point corresponding to serial number i and the sampled point corresponding to serial number i+l it The average value of sum is as 90 percentile.
18. equipment according to claim 11, wherein, the polyplant, for using the coefficient of variation as stability Index, the coefficient of variation of the sampled point of the server of manufacturer to be evaluated is calculated according to below equation,
Coefficient of variation C.V=(SD ÷ MN) × 100%, wherein, averageStandard deviationxjFor some sampled point, n represents the number of sampled point, and j is one in n sampled point The numbering of sampled point.
19. equipment according to claim 11, wherein, the evaluation tool includes fio, MBW, calculates prime number, calculates circle One kind or any combination in frequency, Ubench.
20. equipment according to claim 19, wherein, the performance indications of the fio evaluation tools include disk operating kind Any combination of class/file size species/evaluation metricses species.
21. a kind of server evaluates and tests equipment, wherein, the equipment includes:
Processor;
And it is arranged to store the memory of computer executable instructions, the executable instruction makes the place when executed Manage device:
The sampled point of the server of manufacturer to be evaluated is collected, wherein, the sampled point of collection is the sampled point in preset time, Each sampled point corresponds to a performance indications, and each performance indications corresponds to an evaluation tool;
Sampled point is polymerize, respectively obtains the performance index value and stability indicator of the server of each manufacturer to be evaluated Value, wherein, each performance index value corresponds to a performance indications, and each stability indicator value corresponds to a performance indications;
According to the performance index value and stability indicator value, the comprehensive grading of the server of manufacturer to be evaluated is obtained.
CN201710138866.2A 2016-03-09 2017-03-09 Server evaluating method and equipment Pending CN107391288A (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610131711 2016-03-09
CN2016101317111 2016-03-09

Publications (1)

Publication Number Publication Date
CN107391288A true CN107391288A (en) 2017-11-24

Family

ID=60338303

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710138866.2A Pending CN107391288A (en) 2016-03-09 2017-03-09 Server evaluating method and equipment

Country Status (1)

Country Link
CN (1) CN107391288A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109241096A (en) * 2018-08-01 2019-01-18 北京京东金融科技控股有限公司 Data processing method, device and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103139007A (en) * 2011-12-05 2013-06-05 阿里巴巴集团控股有限公司 Method and system for detecting application server performance
US20140013165A1 (en) * 2007-06-05 2014-01-09 Software Research Inc. Method for System for Testing Websites
CN104361088A (en) * 2014-11-17 2015-02-18 国电南京自动化股份有限公司 Congestion data processing method based on real-time weight analysis in SCADA (supervisory control and data acquisition) system
CN105183604A (en) * 2015-10-29 2015-12-23 浪潮(北京)电子信息产业有限公司 Server performance testing method and system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140013165A1 (en) * 2007-06-05 2014-01-09 Software Research Inc. Method for System for Testing Websites
CN103139007A (en) * 2011-12-05 2013-06-05 阿里巴巴集团控股有限公司 Method and system for detecting application server performance
CN104361088A (en) * 2014-11-17 2015-02-18 国电南京自动化股份有限公司 Congestion data processing method based on real-time weight analysis in SCADA (supervisory control and data acquisition) system
CN105183604A (en) * 2015-10-29 2015-12-23 浪潮(北京)电子信息产业有限公司 Server performance testing method and system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109241096A (en) * 2018-08-01 2019-01-18 北京京东金融科技控股有限公司 Data processing method, device and system

Similar Documents

Publication Publication Date Title
Sileshi A critical review of forest biomass estimation models, common mistakes and corrective measures
Gerke et al. Probing quintessence: reconstruction and parameter estimation from supernovae
US20190018753A1 (en) Software program fault localization
CN109450956B (en) Network security evaluation method, system, medium, and computer system
CN103697928A (en) Instrument calibration method and device
CN107132451A (en) The winding state detection method and system of transformer
CN110347744A (en) Date storage method, device and the equipment of multilayer block chain type account book
CN106528111A (en) Similarity measurement method for data structure job program
D’Urso et al. Scale length does matter: Recommendations for measurement invariance testing with categorical factor analysis and item response theory approaches
CN107391288A (en) Server evaluating method and equipment
Zhu et al. Using Sherman–Morrison theory to remove the coupled effects of multi-transducers in vibration test
CN110362569A (en) The method of calibration and device of tables of data, electronic equipment, storage medium
CN106447144A (en) Target object evaluation method and target object evaluation device
CN105159826B (en) A kind of method and apparatus of wrong sentence in positioning target program
US10713232B2 (en) Efficient data processing
CN108733707A (en) A kind of determining function of search stability and device
Bardsley et al. Optimal design: A computer program to study the best possible spacing of design points for model discrimination
US20230315911A1 (en) Method for data manipulation detection of numerical data values
Haberzettl et al. A new method for studying the longitudinal dynamic behaviour of a suspension on a test rig
Lai et al. High accurate measurement and calibration of the squareness on ultra-precision machine based on error separation
CN114062190B (en) Method, device, terminal and storage medium for characterizing microscopic occurrence characteristics of shale oil
CN113643754B (en) Missense variant gene scoring processing method, optimization scoring method and optimization scoring device
CN112486808B (en) System testing method and device, electronic equipment and storage medium
CN109145996A (en) Achievement data generation method, device and electronic equipment under abnormal environment
CN108763092A (en) A kind of aacode defect detection method and device based on cross validation

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20171124