Disclosure of Invention
The invention aims to provide a health state evaluation method and system based on computer equipment, which solve the problems in the background technology.
In order to solve the technical problems, the invention adopts the following technical scheme that the invention provides a health state assessment method based on computer equipment, which comprises the following steps of ST1, continuously tracking the operation behaviors of students on the computer equipment in a laboratory through a behavior recorder built in a BIOS layer to form a record data set of the operated behaviors of the computers in the laboratory, and continuously monitoring the peripheral operation behaviors of the students on the computer equipment in the laboratory to form a record data set of the peripheral operation behaviors of the computers in the laboratory.
And ST2, recording a data set according to the operated behaviors of the computer in the laboratory and a peripheral data set of the computer in the laboratory, and obtaining the health status grade of a plurality of computer devices in the laboratory through data processing.
And ST3, acquiring teaching related data in a laboratory, performing simulation test on computer equipment in the laboratory, and generating a task abnormality computer equipment set and an energy consumption optimization computer equipment set according to the simulation test data.
And ST4, conveying the task anomaly computer equipment set and the energy consumption optimization computer equipment set to a laboratory equipment management center.
The system for executing the health state assessment method based on the computer equipment comprises a laboratory data acquisition module, a behavior recorder arranged in a BIOS layer, a computer operation record data set and a computer peripheral operation record data set, wherein the behavior recorder is used for continuously tracking the operation behaviors of students on the computer equipment in a laboratory, so that the computer operation record data set is formed, and meanwhile, the peripheral operation behaviors of the students on the computer equipment in the laboratory are continuously monitored, so that the computer peripheral operation record data set is formed.
And the health state evaluation module records a data set according to the operated behaviors of the computer in the laboratory and a peripheral data set of the computer in the laboratory, and obtains health state grades of a plurality of computer devices in the laboratory through data processing.
And the laboratory equipment simulation analysis module is used for acquiring teaching related data in a laboratory, performing simulation test on computer equipment in the laboratory, and generating a task abnormality computer equipment set and an energy consumption optimization computer equipment set according to the simulation test data.
And the processing terminal is used for transmitting the task abnormality computer equipment set and the energy consumption optimization computer equipment set to the laboratory equipment management center.
Aiming at the characteristics of computer equipment in a laboratory, the invention develops a health state evaluation method aiming at the computer equipment in the laboratory, and overcomes the defects in the prior art.
(2) The invention combines the behavior tracing evaluation layer with the peripheral loss quantitative evaluation layer to evaluate the health state of the equipment from the human operation perspective. Physical loss conditions of the peripheral are detected through continuous tracking of the operation behaviors of students on the equipment and accurate detection. And (3) carrying out association analysis on the data of the two to realize the health state assessment of the computer equipment in the laboratory.
(3) According to the invention, the computer equipment is subjected to simulation test aiming at teaching related data in a laboratory, and the scientific research load pressure evaluation layer and the energy consumption behavior evaluation layer are combined to evaluate the performance of the equipment under a high-intensity scientific research task. Meanwhile, the energy consumption condition of the computing equipment under different load states is calculated, and the balance between scientific research requirements and energy-saving targets is realized.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to FIG. 1, a first aspect of the present invention provides a method for evaluating health status based on a computer device, including ST1, continuously tracking operation behaviors of a student on the computer device in a laboratory by a behavior recorder built in a BIOS layer, forming a record data set of the operated behaviors of the computer in the laboratory, and continuously monitoring peripheral operation behaviors of the student on the computer device in the laboratory, forming a record data set of the peripheral operation behaviors of the computer in the laboratory.
In a specific embodiment of the present invention, the laboratory computer operated behavior record data set includes operation logs of a plurality of computer devices used by respective student ID accounts, and the operation logs specifically include a plurality of operation data characteristic values.
Specifically, the operation data characteristic values include, but are not limited to, operation behavior data characteristic values affecting the health state of the computer device, such as a startup frequency, a shutdown frequency, a high-load program running time, a number of simultaneous multi-process starts, a process switching frequency, a number of sleep awakenings, a standby time duty ratio, a magnetic disk continuous read-write time, a GPU full-load running time period, and the like.
In a specific embodiment of the present invention, the peripheral operation behavior record data set of the laboratory computer includes peripheral operation logs used by each student ID account of a plurality of computer devices, and the peripheral operation logs specifically include feature values of a plurality of peripheral operation data under a time sequence feature.
Specifically, the characteristic values of the plurality of peripheral operation data under the time sequence characteristic include, but are not limited to, a capacitance value of each key of the keyboard under the time sequence characteristic, a maximum pressure of the touch screen under the time sequence characteristic, a voltage of the USB interface under the time sequence characteristic, pressing force of a left key and a right key of the mouse under the time sequence characteristic, a resistance coefficient of the roller under the time sequence characteristic and the like, and the characteristic values of the operation behavior data influence the health state of the peripheral of the computer equipment.
The method is characterized in that MCU is utilized to read the keyboard key capacitance value of computer equipment in a laboratory, infrared array scanning touch screen prior pressure distribution is utilized to monitor USB interface voltage by DIY voltage monitoring module, and pressure sensor, microcontroller and encoder are utilized to monitor the pressing force of left and right mouse keys under time sequence characteristics and the resistance coefficient of a roller under time sequence characteristics.
And ST2, recording a data set according to the operated behaviors of the computer in the laboratory and a peripheral data set of the computer in the laboratory, and obtaining the health status grade of a plurality of computer devices in the laboratory through data processing.
In the specific embodiment of the invention, the health status grade of a plurality of computer devices in a laboratory is obtained through data processing, and the specific processing method is that ST2001, a data set is recorded according to the operated behaviors of the computer in the laboratory, and the first health status evaluation value of the plurality of computer devices in the laboratory is evaluated.
ST2002, evaluating second health state evaluation values of a plurality of computer devices in the laboratory according to the computer peripheral data set in the laboratory.
ST2003, carrying out weighted summation processing on the first health state evaluation values and the second health state evaluation values of the plurality of computer devices in the laboratory, and obtaining the overall health state values of the plurality of computer devices in the laboratory through sigmoid function processing.
It should be noted that, through the sigmoid function processing, the overall health state values of a plurality of computer devices in the laboratory can be controlled within the range of 0-1, so that the subsequent analysis and processing are facilitated.
ST2004, comparing the overall health status value of a plurality of computer devices in the laboratory with the overall health status value interval corresponding to each health status level stored in the data warehouse, if the overall health status value of a certain computer device in the laboratory is in the overall health status value interval corresponding to a certain health status level, marking the health status level as the health status level of the computer device, and matching to obtain the health status level of a plurality of computer devices in the laboratory.
It should be noted that, the total health status value interval corresponding to each health status level is specifically set by the laboratory manager, and the higher the health status level is, the larger the corresponding total health status value is.
In a specific embodiment of the invention, the first health state evaluation value of a plurality of computer devices in a laboratory is evaluated, and the specific evaluation method comprises the steps of extracting a plurality of operation data characteristic values of the plurality of computer devices used by each student ID account number from an operation behavior record data set of the computer in the laboratory, comparing the operation data characteristic values with a plurality of operation data characteristic value suitable intervals stored in a data bin, if a certain operation data characteristic value is in the suitable interval, marking the offset parameter of the operation data characteristic value as 0, otherwise, obtaining the exceeding amplitude of the operation data characteristic value, and obtaining the offset parameter of the operation data characteristic value through homogenization treatment.
The above-mentioned exceeding amplitude means an amplitude exceeding the maximum value of the suitable interval or an amplitude exceeding the minimum value of the suitable interval.
And carrying out mean value processing on offset parameters of a plurality of operation data characteristic values of a plurality of computer devices used by each student ID account to obtain first health state evaluation values of the plurality of computer devices.
In a specific embodiment of the invention, the method for evaluating the second health state evaluation value of a plurality of computer devices in a laboratory comprises the steps of extracting characteristic values of a plurality of computer devices used by each student ID account and comprising a plurality of peripheral operation data under time sequence characteristics from a peripheral operation behavior record data set of the computer devices in the laboratory, identifying abnormal time sequence points of the plurality of peripheral operation data, calculating to obtain abnormal proportions of the plurality of peripheral operation data of the plurality of computer devices in the laboratory, if the abnormal proportions of the peripheral operation data are greater than or equal to preset abnormal proportions, marking the second health state evaluation value of the computer devices in the laboratory as 0, and if the abnormal proportions of all the peripheral operation data are smaller than the preset abnormal proportions, marking the second health state evaluation value of the computer devices as 1.
The abnormal time sequence point of the plurality of peripheral operation data is identified, specifically, the abnormal time sequence point is identified based on the characteristic value of the plurality of peripheral operation data under the time sequence characteristic and the proper characteristic value interval of the plurality of peripheral operation data stored in the data bin, if the characteristic value of the peripheral operation data at a certain time sequence point is not within the proper characteristic value interval, the time sequence point is marked as the abnormal time sequence point, and the proper characteristic value interval of the plurality of peripheral operation data stored in the data bin is specifically formulated according to an operation manual or an industry standard of the peripheral equipment.
The method comprises the steps of counting the number of abnormal time sequence points of the peripheral operation data when the plurality of computer devices in the laboratory are used by the student ID accounts, dividing the number of the abnormal time sequence points by the total number of the time sequence points to obtain the abnormal proportion of the peripheral operation data when the plurality of computer devices in the laboratory are used by the student ID accounts, and carrying out mean value processing on the abnormal proportion to obtain the abnormal proportion of the peripheral operation data of the plurality of computer devices in the laboratory.
The invention combines the behavior tracing evaluation layer with the peripheral loss quantitative evaluation layer to evaluate the health state of the equipment from the human operation perspective. Physical loss conditions of the peripheral are detected through continuous tracking of the operation behaviors of students on the equipment and accurate detection. And (3) carrying out association analysis on the data of the two to realize the health state assessment of the computer equipment in the laboratory.
And ST3, acquiring teaching related data in a laboratory, performing simulation test on computer equipment in the laboratory, and generating a task abnormality computer equipment set and an energy consumption optimization computer equipment set according to the simulation test data.
In the specific embodiment of the invention, the method comprises the steps of ST3011, screening the required health state level of the computer equipment in the laboratory according to the teaching related data in the laboratory, rejecting a plurality of computer equipment lower than the required health state level in the laboratory, and recording the rest of computer equipment as a plurality of test computers.
It should be noted that, according to the teaching related data in the laboratory, the computer equipment demand health status level in the storage laboratory is screened, specifically, a plurality of key parameters related to the computer equipment pressure are extracted, the demand health status level corresponding to each key parameter preset in the data bin is related to the demand health status level corresponding to a plurality of key parameters related to the computer equipment pressure, and the maximum demand health status level is selected as the computer equipment demand health status level in the laboratory.
ST3012, according to the relevant data of teaching in the laboratory, carry out the simulation test to a plurality of remaining computer equipment, the simulation test is specifically according to the relevant data of teaching in the laboratory to obtain the simulation test data, the simulation test data specifically includes the characteristic parameter of a plurality of test computers in each test in the laboratory.
Illustratively, from teaching-related data within a laboratory (e.g., course outline, student experiment report, historical equipment monitoring log, etc.), key parameters related to computer equipment pressure are extracted to form a test "benchmark template".
Taking a parallel computing scene, a data IO scene and a graph rendering scene as examples, a reference template of the parallel computing scene is the thread number of a multithreaded task, the number of concurrent processes, the task type, a reference template of the data IO scene is the data set size of SQL query, the batch read-write frequency of file operation, a read-write mode, the reference template of the graph rendering scene is the rendering resolution, the number of polygons of a 3D model and the rendering time.
It should be noted that, simulation of the parallel computing scene may be implemented by Python, simulation of the data IO scene may be implemented by Linux, and simulation of the graphics rendering scene may be implemented by Blender, adobe AFTER EFFECTS.
Specifically, the characteristic parameters of each test include a test type, a fault type and a total power consumption, wherein the fault type includes but is not limited to a fault type of a parallel computing scene, such as insufficient CPU utilization rate, abnormal temperature, task completion delay, CPU frequency reduction, program breakdown, abnormal writing speed of a data IO scene, abnormal hard disk temperature, insufficient GPU utilization rate of a graphic rendering scene, occupied video memory and the like, under various test scenes.
ST3021, extract the fault type of several test computer devices in the laboratory in each test from the simulated test data, count and calculate and get the frequency of the various faults of several test computer devices, if the frequency of the fault of a certain test computer device is greater than or equal to the frequency threshold of the fault of the change type stored in the data warehouse, then the test computer is summarized into the task abnormal computer device set A, and according to the rule codes of several fault types stored in the data warehouse, the display rule codes of the test computer in the task abnormal computer device set are generated.
For example, if a certain test computer is numbered 4, a certain fault type is insufficient in CPU utilization, and the recurrence frequency of the fault type is greater than or equal to the corresponding recurrence frequency threshold, the rule code of the fault type is extracted from the data bin to be 1U, and if the rule code displayed by the test computer device in the energy consumption optimizing computer device set is。
ST3022, extracting total power consumption and test types of a plurality of test computer devices in a laboratory in each test from simulation test data, extracting allowable power consumption of a plurality of test types from a data warehouse, if the total power consumption of a certain test exceeds the corresponding allowable power consumption, inducing the test computer devices into an energy consumption optimizing computer device set B, and generating display rule codes of the test computer in the energy consumption optimizing computer device set according to the test types.
For example, if a test computer is numbered 5, a test type is a parallel computing scenario type, the total power consumption of the test exceeds the corresponding allowable power consumption, and the rule code of the test type extracted from the data bin is 0I, the rule code displayed by the test computer device in the energy consumption optimizing computer device set is。
ST3023, the task anomaly computer device set A and the energy consumption optimization computer device set B satisfy the following conditions ifThen,The first bit rule code of an element in the set of computer devices is optimized for energy consumption.
It is to be added that if the first rule code of a certain element in the energy consumption optimizing computer equipment set is the same as the first rule code of a certain element in the task abnormality computer equipment set, the element is removed from the energy consumption optimizing computer equipment set.
It should be noted again that, the frequency threshold value of the recurrence of each fault and the allowable power consumption of a plurality of test types stored in the data warehouse are specifically set by the laboratory manager, and the laboratory manager can set up pertinently according to the laboratory requirement without limitation.
According to the invention, the computer equipment is subjected to simulation test aiming at teaching related data in a laboratory, and the scientific research load pressure evaluation layer and the energy consumption behavior evaluation layer are combined to evaluate the performance of the equipment under a high-intensity scientific research task. Meanwhile, the energy consumption condition of the computing equipment under different load states is calculated, and the balance between scientific research requirements and energy-saving targets is realized.
And ST4, conveying the task anomaly computer equipment set and the energy consumption optimization computer equipment set to a laboratory equipment management center.
Aiming at the characteristics of computer equipment in a laboratory, the invention develops a health state evaluation method aiming at the computer equipment in the laboratory, and overcomes the defects in the prior art.
Referring to fig. 2, a second aspect of the present invention provides a system for the health status assessment method based on a computer device according to the present invention, which includes a laboratory data acquisition module, through a behavior recorder built in a BIOS layer, continuously tracking the operation behavior of a student on the computer device in the laboratory, to form a record data set of the operated behavior of the computer in the laboratory, and simultaneously continuously monitoring the peripheral operation behavior of the student on the computer device in the laboratory, to form a record data set of the peripheral operation behavior of the computer in the laboratory.
And the health state evaluation module records a data set according to the operated behaviors of the computer in the laboratory and a peripheral data set of the computer in the laboratory, and obtains health state grades of a plurality of computer devices in the laboratory through data processing.
And the laboratory equipment simulation analysis module is used for acquiring teaching related data in a laboratory, performing simulation test on computer equipment in the laboratory, and generating a task abnormality computer equipment set and an energy consumption optimization computer equipment set according to the simulation test data.
And the processing terminal is used for transmitting the task abnormality computer equipment set and the energy consumption optimization computer equipment set to the laboratory equipment management center.
It should be noted that the present invention also includes a data bin for storing the relevant data required by the present invention.
The foregoing is merely illustrative and explanatory of the principles of this invention, as various modifications and additions may be made to the specific embodiments described, or similar arrangements may be substituted by those skilled in the art, without departing from the principles of this invention or beyond the scope of this invention as defined in the claims.