CN116647465A - Automatic operation and maintenance method based on artificial intelligence technology - Google Patents
Automatic operation and maintenance method based on artificial intelligence technology Download PDFInfo
- Publication number
- CN116647465A CN116647465A CN202310654748.2A CN202310654748A CN116647465A CN 116647465 A CN116647465 A CN 116647465A CN 202310654748 A CN202310654748 A CN 202310654748A CN 116647465 A CN116647465 A CN 116647465A
- Authority
- CN
- China
- Prior art keywords
- target monitoring
- asset
- monitoring object
- maintenance
- artificial intelligence
- 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
Links
- 238000012423 maintenance Methods 0.000 title claims abstract description 58
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000005516 engineering process Methods 0.000 title claims abstract description 18
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 17
- 238000012544 monitoring process Methods 0.000 claims abstract description 100
- 230000002159 abnormal effect Effects 0.000 claims abstract description 25
- 238000013507 mapping Methods 0.000 claims abstract description 10
- 238000007689 inspection Methods 0.000 claims abstract description 9
- 230000008520 organization Effects 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims description 21
- 238000001514 detection method Methods 0.000 claims description 18
- 230000000694 effects Effects 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 12
- 238000013508 migration Methods 0.000 claims description 6
- 230000005012 migration Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000013475 authorization Methods 0.000 claims description 4
- 238000012790 confirmation Methods 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 abstract description 3
- 230000005540 biological transmission Effects 0.000 description 5
- 230000004044 response Effects 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 3
- 238000012163 sequencing technique Methods 0.000 description 3
- 241001522296 Erithacus rubecula Species 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000004148 unit process Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/149—Network analysis or design for prediction of maintenance
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention discloses an automatic operation and maintenance method based on an artificial intelligence technology, and relates to the technical field of intelligent operation and maintenance; the method comprises the following steps: automatically acquiring asset information of the current organization or the enterprise internal network; establishing an asset mapping relation and a target monitoring object list; automatically inspecting the target monitoring object; intelligent judgment and attribution are carried out on the inspection abnormal results, and abnormal treatment is carried out on reason attribution; and analyzing and predicting according to the recent operation and maintenance data to obtain abnormal conditions of the resources in a certain time interval in the future, and establishing a solution plan in advance. According to different standards, a plurality of target monitoring object lists are established, an adaptive list is selected, and monitoring time is dynamically adjusted when assets in the list are monitored, so that the whole operation and maintenance system is more intelligent, the working states of all current network assets are matched in real time, and resource tilting is carried out in a targeted mode, and the operation and maintenance system is more efficient and reasonable.
Description
Technical Field
The invention relates to the technical field of intelligent operation and maintenance, in particular to an automatic operation and maintenance method based on an artificial intelligence technology.
Background
The operation and maintenance generally refers to the maintenance of network software and hardware which is established by a large organization, wherein the traditional operation and maintenance refers to the information technology operation and maintenance, namely I T operation and maintenance.
With the continuous development of the internet, the scale of internet companies is continuously expanded, no matter manpower or machine resources are rapidly increased, operation and maintenance work in the internet companies is continuously expanded, if a traditional manual operation and maintenance mode is adopted, operation and maintenance timeliness is easily lost, the operation error rate is increased, and the requirement of rapid iteration of the system structure internetworking cannot be met.
Under the support of artificial intelligence technology and big data, system maintenance is carried out on a large number of equipment by utilizing an automatic operation script, so that unified monitoring and unified operation and maintenance scheduling are realized; however, the existing automatic operation and maintenance is performed according to manual setting rules, deployment is performed after equipment faults are detected or configuration is updated according to the rules, the existing automatic operation and maintenance is equivalent to semi-manual operation, the intelligent operation and maintenance is not intelligent, and the faults cannot be perceived and early warned in advance; meanwhile, an information island is formed between the single devices, so that comprehensive analysis and processing of a large number of demands and feedback cannot be performed, the operation efficiency is reduced, and information resources are wasted; to this end, we provide an automated operation and maintenance method based on artificial intelligence technology to solve the above-mentioned problems.
Disclosure of Invention
The invention aims to provide an automatic operation and maintenance method based on an artificial intelligence technology.
The technical problems solved by the invention are as follows:
(1) How to collect asset information by combining an active detection mode and a passive detection mode, and grouping the asset information according to different application scenes and standards to establish a target monitoring object list and an asset mapping relation, so that the problem that in the prior art, adaptive selection cannot be carried out according to the application scenes is solved;
(2) How to automatically patrol according to the selected target monitoring object list, different monitoring time allocation strategies are adopted in the initialization and subsequent round of patrol periods respectively, so that the problem that the distribution can only be carried out according to the set monitoring time in the prior art and cannot be adjusted in real time according to actual conditions is solved;
(3) How to identify, judge and attributing the collected data, and formulate a targeted solving strategy to finally construct a case base, thereby solving the problem that the emergency situation cannot be effectively handled in time in the prior art.
The invention can be realized by the following technical scheme: an automatic operation and maintenance method based on an artificial intelligence technology comprises the following steps:
step one, automatically acquiring asset information of the current organization or an enterprise internal network;
establishing an asset mapping relation and a target monitoring object list;
step three, automatically inspecting the target monitoring object;
step four, intelligent judgment and attribution are carried out on the inspection abnormal results, and abnormal treatment is carried out on reason attribution;
fifthly, analyzing and predicting to obtain abnormal conditions of resources in a certain time interval in the future according to recent operation and maintenance data, and establishing a solution plan in advance;
step six, constructing a case library in the storage unit, and automatically matching when an emergency abnormal situation occurs, so that the abnormal processing is completed rapidly.
The invention further technically improves that: in the first step, two modes of active detection and passive detection are adopted to automatically acquire asset information, and when an operation server is started each time, the passive detection mode is to detect flow data of nodes of the network at fixed time through detectors deployed at network nodes, and keyword matching is carried out to acquire asset information of newly added assets.
The invention further technically improves that: in the second step, the assets are respectively grouped according to four aspects of standards of functions, service department blocks, asset value and flow data, a target monitoring object list and asset mapping relations in the list are established, and in different application scenes, different target monitoring object lists are selected to monitor the assets.
The invention further technically improves that: in the third step, each target monitoring object is automatically and circularly patrolled according to the selected target monitoring object list, and under the initialization condition, the patrol times and duration of different target monitoring objects and the classification of network resources are distributed according to the sequence on the target monitoring object list, and the higher specific gravity and the later sequencing are carried out;
after the initialization patrol is completed, the monitoring time of each target monitoring object in the round patrol period is redistributed according to the combination of the active times and the asset value of the target monitoring object.
The invention further technically improves that: the specific operation of reassigning the monitoring time of each target monitoring object in the round period comprises the following steps:
the whole round period is set as T, the number of target monitoring objects is N, and the liveness of the target monitoring objects is respectively marked as Ac i The asset value of the target monitoring object is respectively marked as V i Wherein, the method comprises the steps of, wherein, i=1, 2,3. Once again, the total number of the components, substituted into the following formula:
wherein t is i The monitoring time allocated by the corresponding target monitoring object is represented, alpha and beta are respectively weight compensation coefficients of the activity degree and the asset value of the target monitoring object, alpha and beta are fixed values, and alpha is<Beta; ζ is the compensation coefficient for the trim equation.
The invention further technically improves that: in the third step, the operation and maintenance server performs real-time identification and judgment on the acquired data, and correspondingly generates a storage space alarm signal, a resource exhaustion signal, a network bandwidth mismatch signal and a version error signal when the acquired data exceeds a set interval of the data.
The invention further technically improves that: in step four, attribution processing is performed according to the plurality of abnormal signals generated in step three, specifically:
when a storage space alarm signal is received, automatically clearing the storage space garbage, monitoring a long-time storage file with low activity of the storage space, automatically deleting the file after confirmation authorization of an actual user is obtained when the file is not active in the monitoring time, and sending an instruction to inform operation and maintenance personnel to expand the capacity when the storage space is insufficient after the processing;
after receiving the resource exhaustion signal, counting the overrun times of the recent resource utilization rate and the average utilization rate of the asset, and comparing the threshold value, and upgrading the asset when the performance of the asset is insufficient to match the requirement;
when a network bandwidth mismatch signal is received, adjusting the network bandwidth to enable the ratio of the network bandwidth to the network load of the corresponding equipment to be within [ a, b ];
and after receiving the version error signal, updating the version of the asset in a silent configuration mode.
The invention further technically improves that: in the fifth step, a data migration curve is established according to all data of the corresponding assets in the storage unit, a function is fitted, a first derivative and a second derivative are solved for the function, curve trend in a future period of time is predicted from positive and negative values of the first derivative and the second derivative, and a plan is formulated.
Compared with the prior art, the invention has the following beneficial effects:
1. the asset information is collected by combining an active detection mode and a passive detection mode, and grouping is carried out according to different application scenes and standards, so that a target monitoring object list and an asset mapping relation are established, and in different application scenes, an adaptive target monitoring object can be selected, and further, the method has tendencies and pertinence in different asset operation and maintenance logics, and is favorable for carrying out a timely and efficient operation and maintenance method on important assets;
2. according to the selected target monitoring object list, carrying out automatic inspection, adopting different monitoring time allocation strategies in the initialization period and the subsequent round inspection period respectively, and carrying out allocation according to the sequence in the target monitoring object list in the initialization period, wherein the allocation occupation ratio of the allocation which is arranged at the front is large; in the subsequent round of inspection period, according to the calculation result of the monitoring time combined with the activity and the asset value, and the monitoring time is dynamically adjusted under the adjustment of the grade frame, the automatic dynamic adjustment according to the working state of each target monitoring object is ensured, the whole operation and maintenance system is more intelligent, the working states of all the current network assets are matched in real time to conduct resource inclination in a targeted manner, and the operation and maintenance system is more efficient and reasonable;
3. the method also utilizes the existing data to establish a data migration curve and analyze the walking trend of the data, thereby predicting the monitoring data condition of a period in the future, formulating a plan in a targeted manner and effectively reducing or preventing the occurrence of the abnormal condition.
Drawings
The present invention is further described below with reference to the accompanying drawings for the convenience of understanding by those skilled in the art.
FIG. 1 is a schematic flow chart of an automated operation and maintenance method of the present invention;
fig. 2 is a system block diagram of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention for achieving the preset aim, the following detailed description is given below of the specific implementation, structure, characteristics and effects according to the invention with reference to the attached drawings and the preferred embodiment.
Referring to fig. 1-2, the automated operation and maintenance method based on the artificial intelligence technology comprises the following steps:
s1, automatically acquiring assets in a current organization or an enterprise internal network;
s2, establishing an asset mapping relation and a target monitoring object list;
s3, automatically inspecting the target monitoring object;
s4, intelligent judgment and attribution are carried out on the inspection abnormal results, and abnormal treatment is carried out on reason attribution;
s5, analyzing and predicting to obtain abnormal conditions of the resources in a certain time interval in the future according to the recent operation and data.
Aiming at the step S1, the assets comprise a server, a router, a switch, an application program, an operating system, a printer, a computer and other terminal equipment, and an asset detection unit is arranged in an operation and maintenance server and comprises an active detection mode and a passive detection mode;
the active detection mode sends a specific data access packet to the assets in all access networks, wherein the data access packet is mainly a hypertext transfer protocol (HTTP), and the target asset receiving the data access packet returns an HTTP response data packet, so that the asset detection unit can acquire the equipment brand, model, port, operating system, application service information, equipment type related fingerprint information and the like of the corresponding asset from a response head and a response body of the HTTP response data packet; when restarting the operation server each time, acquiring asset information in an active detection mode once;
the passive detection mode is to detect flow data flowing through a corresponding network node at regular time by arranging a detector at the network node, extract data from the flow data, and match the keywords of fields in the corresponding data, in this embodiment, we use an N-Gram model to match the keywords, so as to obtain asset information of newly added assets; the asset information acquired in any mode is updated to the storage unit;
aiming at step S2, complex relations among network assets exist, including physical connection, logical connection and data flow direction, and the asset types are different, so that when an asset mapping relation is established, the assets need to be distinguished according to different standards and rules;
dividing according to functions, dividing assets with the same type of functions into the same group, such as data storage, data processing, network relay transmission, service execution terminals (including printing, displaying and the like);
dividing according to the business department plate, dividing the assets of the business scope of the same function into the same group (such as a technical function module, an administrative function module or a sales function module, etc.);
dividing the asset value according to the asset value grade, wherein the asset value is in enterprise level, department level and individual level, the enterprise level assets comprise enterprise databases, enterprise level servers and system assets with high equipment value and information value of firewall and enterprise level application software such as OA, CRM and the like, and the department level assets comprise gateway equipment such as router switches and the like and department level application software such as the software developed by departments, the special application software of departments and the like;
dividing according to the transmission size of the flow data and the number of the connection nodes, distinguishing the abnormal processing priority of the assets, advancing the priority of the assets with large transmission amount of the flow data and large number of the connection nodes, carrying out normalization processing on the transmission amount of the flow data and the number of the connection nodes, setting weight values of the transmission amount of the flow data and the number of the connection nodes in influencing service development, carrying out summation after reassigning according to the weight values to obtain influence weight points, arranging the influence weight points according to the sequence from large to small, and determining the generated ordered list as a priority list;
obtaining a plurality of target monitoring object lists and asset mapping relations in the lists according to the plurality of standards or rules;
aiming at step S3, according to different organization structures and application scenes of corresponding enterprises, selecting a target monitoring object list matched with the corresponding enterprise to automatically patrol, and according to the change of the application scene, replacing other lists again; in the process of initializing and inspecting, because the network memory resources or calculation power is limited, the inspection times, the duration and the distribution of the network resources are inclined towards the front target asset in the target monitoring object list;
when the system automatically patrols, one or more network management protocols in SNMP, WMI or Telnet/SSH can be selected for carrying out uninterrupted state round patrol on the target monitoring objects so as to monitor the working state of each target monitoring object;
specifically, setting a round robin period, and in the round robin period, performing data acquisition on disk space, CPU (central processing unit) utilization rate, memory utilization rate, network bandwidth, system version and activity of each target monitoring object and storing the data into a storage unit of an operation and maintenance server;
meanwhile, a data judging unit of the operation and maintenance server carries out real-time identification judgment on the collected data:
when the disk space of the corresponding asset is less than 20%, judging that the hard disk storage space is tense, and generating a storage space alarm signal;
when the CPU utilization rate or the memory utilization rate of the corresponding asset exceeds 80%, judging that the corresponding asset runs on a card and the computing power resources are exhausted, and generating a resource exhaustion signal;
under the condition that the influence of a network topological structure is not considered, carrying out ratio operation on the network bandwidth and the network load of the corresponding equipment, judging that the network bandwidth is normal when the ratio result is between [ a, b ], otherwise, generating a network bandwidth mismatch signal, wherein a and b are preset values and 0< a <1< b;
comparing the system version number of the corresponding asset with the system version number recorded in the operation server, when the system version number is matched and consistent, indicating that the current system version is correct, and otherwise, generating a version error signal;
in the round-robin period, each target monitoring object is distributed to a continuous monitoring time, and the active times of the target monitoring objects are acquired in the corresponding continuous monitoring time, wherein the active times comprise network request times and hardware use times; performing ratio operation on the number of active times of the target monitoring object and the monitoring time for completing the number of active times, so as to obtain the activity, and sequencing according to the activity, so as to redistribute the monitoring time of each target monitoring object in the round period according to the sequencing condition of the activity and combining the value of the corresponding asset:
specifically, let the whole round period be T, the number of target monitoring objects be N, and the liveness of the target monitoring objects be Ac i The asset value of the target monitoring object is respectively marked as V i Wherein, the method comprises the steps of, wherein, i=1, 2,3. Once again, the total number of the components, substituting into a monitoring time allocation formula:
wherein t is i The monitoring time allocated by the corresponding target monitoring object is represented, alpha and beta are respectively weight compensation coefficients of the activity degree and the asset value of the target monitoring object, alpha and beta are fixed values, and alpha is<β；
And is also provided withWherein ζ is a compensation coefficient for balancing the equation;
after the monitoring time obtained by the allocation of each target monitoring object is obtained, each target monitoring object is monitored in the next round of polling period according to the calculation result, so that the automatic dynamic adjustment according to the working state of each target monitoring object is ensured, the whole operation and maintenance system is more intelligent, the working states of all current network assets are matched in real time, and the resource inclination is carried out, so that the operation and maintenance system is more efficient and reasonable;
it should be noted that, the monitoring time ordering of the dynamically adjusted target monitoring objects needs to be performed under the level frame of the current target monitoring object list, that is, the crossing between the level groups cannot be performed, that is, a monitoring time threshold is set in each level, and the calculation result is adjusted by the threshold.
For step S4, an attribution processing unit for performing attribution and abnormality handling is provided in the operation and maintenance server, and the attribution processing unit performs corresponding processing after receiving the plurality of types of abnormality signals of the data judgment unit, specifically:
when a storage space alarm signal is received, automatically starting garbage cleaning, and traversing files in the storage space to obtain the storage time and the size of the files in the storage space, counting the use times of the files, marking the files which exceed a set time length and are lower than the set use times, and generating a to-be-cleaned form by the marked files, wherein the contents of the form comprise the size, the storage time and the use times of the files; continuously monitoring files in the form to be cleaned, deleting the files from the form to be cleaned when the files are re-active in a set monitoring period, transmitting the form to be cleaned to actual users of corresponding assets for confirmation authorization after the monitoring period is exceeded, and directly and automatically deleting the corresponding files of the storage space after the confirmation authorization is obtained so as to ensure the residual capacity of the storage space;
in addition, when the deletable file is not available or the storage space is insufficient after the file is deleted, a capacity expansion instruction is directly generated to corresponding operation and maintenance personnel for capacity expansion;
when the resource exhaustion signal is received, the recent data of the corresponding asset is pulled from the storage unit, the number of times that the recent resource utilization rate of the corresponding asset exceeds the set value and the recent average resource utilization rate are counted, and when the number of times exceeds/rounds is countedAnd average resource usage rate>When 60%, considering that the performance of the current asset is insufficient to match the service requirement, and updating is needed, and sending an updating signal to corresponding operation and maintenance personnel;
when a network bandwidth mismatch signal is received, adjusting the network bandwidth to enable the ratio of the network bandwidth to the network load of the corresponding equipment to be within [ a, b ];
and after receiving the version error signal, searching the version number of the system recorded in the operation and maintenance server to configure the corresponding version update package, wherein the configuration mode adopts silent configuration, and the normal use of the corresponding asset is not influenced in the configuration process.
For step S5, all data of a time node to a set time interval at the current moment are acquired from a storage unit, analysis processing is carried out on the data of each target monitoring object, data migration curves of the target monitoring objects in a plurality of round-robin periods are continuously drawn, corresponding data migration curves are fitted by utilizing a functional relation, the first derivative and the second derivative are solved according to the functional relation, then curve trend in a future period of time from the current round-robin period is judged according to positive and negative signs of the first derivative and the second derivative, working state data of the target monitoring objects corresponding to a future moment is predicted according to the curve trend, faults and potential anomaly hazards which possibly occur are found in advance, and accordingly a plan is formulated, and a solution is formed in a targeted manner;
more specifically, after the processing unit processes the abnormal signal, the cause of the abnormality of the current target monitoring object and the processing scheme are recorded, an abnormal case library is set in a storage unit in a open space, when the unexpected abnormal situation is handled and the operation and maintenance period is tense, the specific data migration curve forming the abnormal situation corresponds to the processing case when the same type of problem occurs in the case library when the same type of asset is matched, the processing is performed according to the matched processing case, the processing capability of the sudden abnormal situation is greatly improved, and the whole operation and maintenance system has the machine learning capability.
The present invention is not limited to the above embodiments, but is capable of modification and variation in all aspects, including those of ordinary skill in the art, without departing from the spirit and scope of the present invention.
Claims (9)
1. The automatic operation and maintenance method based on the artificial intelligence technology is characterized by comprising the following steps of:
step one, automatically acquiring asset information of the current organization or an enterprise internal network;
establishing an asset mapping relation and a target monitoring object list;
step three, automatically inspecting the target monitoring object;
step four, intelligent judgment and attribution are carried out on the inspection abnormal results, and abnormal treatment is carried out on reason attribution;
fifthly, analyzing and predicting to obtain abnormal conditions of resources in a certain time interval in the future according to recent operation and maintenance data, and establishing a solution plan in advance;
step six, constructing a case library in the storage unit, and automatically matching when an emergency abnormal situation occurs, so that the abnormal processing is completed rapidly.
2. The automated operation and maintenance method based on artificial intelligence technology according to claim 1, wherein in the first step, asset information is automatically acquired by adopting two modes of active detection and passive detection, the acquisition time of the active detection mode is each time the operation and maintenance server is started, and the passive detection mode is to detect flow data of nodes of the network at regular time through detectors deployed at the network nodes, and perform keyword matching to acquire asset information of newly added assets.
3. The automated operation and maintenance method based on artificial intelligence technology according to claim 1, wherein in the second step, the assets are respectively grouped according to four aspect standards of functions, service department blocks, asset value and flow data, a target monitoring object list and asset mapping relations in the list are established, and in different application scenarios, different target monitoring object lists are selected to monitor the assets.
4. The automated operation and maintenance method based on artificial intelligence technology according to claim 1, wherein in the third step, each target monitoring object is automatically and alternately patrolled according to the selected target monitoring object list, and under the initialization condition, the number of patrols and duration of different target monitoring objects and the classification of network resources are distributed according to the sequence on the target monitoring object list, and the ratio of the first ordered objects to the second ordered objects is high;
and in the round period after the initialization and the inspection are completed, dynamically adjusting the monitoring time of each target monitoring object in the round period according to the combination of the active times and the asset value of the target monitoring object.
5. The automated operation and maintenance method based on artificial intelligence technology according to claim 4, wherein the specific operation of obtaining the dynamically adjusted monitoring time comprises:
the whole round period is set as T, the number of target monitoring objects is N, and the liveness of the target monitoring objects is respectively marked as Ac i The asset value of the target monitoring object is respectively marked as V i Wherein, the method comprises the steps of, wherein, i=1, 2,3. Once again, the total number of the components, substituted into the following formula:
wherein t is i The monitoring time allocated by the corresponding target monitoring object is represented, alpha and beta are respectively weight compensation coefficients of the activity degree and the asset value of the target monitoring object, alpha and beta are fixed values, and alpha is<Beta; ζ is the compensation coefficient for the trim equation.
6. The automated operation and maintenance method based on artificial intelligence technology according to claim 1, wherein in the third step, the operation and maintenance server performs real-time identification and judgment on the collected data, and generates a storage space alarm signal, a resource exhaustion signal, a network bandwidth mismatch signal and a version error signal when the collected data exceeds a set interval of the data.
7. The automated operation and maintenance method based on artificial intelligence technology according to claim 6, wherein in step four, attribution processing is performed according to the plurality of abnormal signals generated in step three, specifically:
when a storage space alarm signal is received, automatically clearing the storage space garbage, monitoring a long-time storage file with low activity of the storage space, automatically deleting the file after confirmation authorization of an actual user is obtained when the file is not active in the monitoring time, and sending an instruction to inform operation and maintenance personnel to expand the capacity when the storage space is insufficient after the processing;
after receiving the resource exhaustion signal, counting the overrun times of the recent resource utilization rate and the average utilization rate of the asset, and comparing the threshold value, and upgrading the asset when the performance of the asset is insufficient to match the requirement;
when a network bandwidth mismatch signal is received, adjusting the network bandwidth to enable the ratio of the network bandwidth to the network load of the corresponding equipment to be within [ a, b ];
and after receiving the version error signal, updating the version of the asset in a silent configuration mode.
8. The automated operation and maintenance method based on artificial intelligence technology according to claim 1, wherein in step five, a data migration curve is established according to all data of the corresponding asset in the storage unit and a function is fitted, a first derivative and a second derivative are solved for the function, curve trend in a future period is predicted from positive and negative signs of the first derivative and the second derivative, and a plan is formulated.
9. The automated operation and maintenance method according to claim 5, wherein the dynamically adjusting the monitoring time sequence is performed under a hierarchical framework of a current target monitoring object list, that is, cannot be performed across hierarchical groupings, and a monitoring time threshold is set between hierarchical groupings, and the calculation result is adjusted by the threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310654748.2A CN116647465A (en) | 2023-06-05 | 2023-06-05 | Automatic operation and maintenance method based on artificial intelligence technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310654748.2A CN116647465A (en) | 2023-06-05 | 2023-06-05 | Automatic operation and maintenance method based on artificial intelligence technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116647465A true CN116647465A (en) | 2023-08-25 |
Family
ID=87619669
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310654748.2A Pending CN116647465A (en) | 2023-06-05 | 2023-06-05 | Automatic operation and maintenance method based on artificial intelligence technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116647465A (en) |
-
2023
- 2023-06-05 CN CN202310654748.2A patent/CN116647465A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7720841B2 (en) | Model-based self-optimizing distributed information management | |
| CN107707376B (en) | A kind of method and system of monitoring and alarm | |
| US9672085B2 (en) | Adaptive fault diagnosis | |
| US10318366B2 (en) | System and method for relationship based root cause recommendation | |
| Qu et al. | A new dependency and correlation analysis for features | |
| US6658367B2 (en) | System for time-bucketing of baselined data collector data | |
| US10909018B2 (en) | System and method for end-to-end application root cause recommendation | |
| US20020124078A1 (en) | System for self-monitoring of SNMP data collection process | |
| CN102739802A (en) | Service application-oriented IT contralized operation and maintenance analyzing system | |
| US20130212257A1 (en) | Computer program and monitoring apparatus | |
| CN103081407A (en) | Failure analysis device, and system and method for same | |
| CN106656682A (en) | Method, system and device for detecting cluster heartbeat | |
| CN103795575B (en) | A kind of system monitoring method towards multiple data centers | |
| WO2017076256A1 (en) | Load balancing method, related device and system | |
| El-Shamy et al. | Anomaly detection and bottleneck identification of the distributed application in cloud data center using software–defined networking | |
| CN112769605A (en) | Heterogeneous multi-cloud operation and maintenance management method and hybrid cloud platform | |
| CN114553747A (en) | Method, device, terminal and storage medium for detecting abnormality of redis cluster | |
| CN117221088A (en) | Computer network intensity detection system and device | |
| CN116647465A (en) | Automatic operation and maintenance method based on artificial intelligence technology | |
| CN106210120B (en) | A kind of recommended method and its device of server | |
| CN106341474B (en) | It is a kind of that center and its contents management method are managed based on the data of ICN and SDN network | |
| KR100500836B1 (en) | Fault management system of metro ethernet network and method thereof | |
| CN111988172B (en) | Network information management platform, device and security management method | |
| CN113300914A (en) | Network quality monitoring method, device, system, electronic equipment and storage medium | |
| CN106888237A (en) | A kind of data dispatching method and system |
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 |