CN117472511A - Container resource monitoring method, device, computer equipment and storage medium - Google Patents

Container resource monitoring method, device, computer equipment and storage medium Download PDF

Info

Publication number
CN117472511A
CN117472511A CN202311489628.8A CN202311489628A CN117472511A CN 117472511 A CN117472511 A CN 117472511A CN 202311489628 A CN202311489628 A CN 202311489628A CN 117472511 A CN117472511 A CN 117472511A
Authority
CN
China
Prior art keywords
peak
information
period
trend
target
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
CN202311489628.8A
Other languages
Chinese (zh)
Inventor
周智涵
吴延生
张晓鹏
周新衡
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.)
Industrial and Commercial Bank of China Ltd ICBC
Original Assignee
Industrial and Commercial Bank of China Ltd ICBC
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 Industrial and Commercial Bank of China Ltd ICBC filed Critical Industrial and Commercial Bank of China Ltd ICBC
Priority to CN202311489628.8A priority Critical patent/CN117472511A/en
Publication of CN117472511A publication Critical patent/CN117472511A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/30Monitoring
    • G06F11/3003Monitoring arrangements specially adapted to the computing system or computing system component being monitored
    • G06F11/3006Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/044Recurrent networks, e.g. Hopfield networks
    • G06N3/0442Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • G06N3/084Backpropagation, e.g. using gradient descent
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • G06F2009/45591Monitoring or debugging support
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • G06F2009/45595Network integration; Enabling network access in virtual machine instances
    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Quality & Reliability (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application relates to a container resource monitoring method, a device, computer equipment, a storage medium and a computer program product, and relates to the technical fields of financial science and technology and artificial intelligence. The method comprises the following steps: acquiring container resource use information of a business object in a business handling system; training the long-period memory network model according to the historical use information to obtain a first trend prediction model, and training the long-period memory network model according to the real-time use information to obtain a second trend prediction model; under the condition that the target period prediction information meets a preset alarm threshold value, determining a peak period initiation type of a service peak in the target period in the current day according to the current day overall prediction information; and outputting a processing measure prompt message matched with the peak time initiation type. By adopting the method, the operation efficiency of the business handling system can be improved.

Description

Container resource monitoring method, device, computer equipment and storage medium
Technical Field
The present application relates to the field of artificial intelligence, and in particular, to a method, an apparatus, a computer device, a storage medium, and a computer program product for monitoring container resources.
Background
With the rapid development of the financial industry, the iteration speed of the version of the business handling system used by the business object is increased, the number of users of the business object is increased gradually, the business volume is increased gradually, the abnormal situation of the business handling system is increased, and the influence on the business handling process caused by the abnormality is greater.
When the service handling system is abnormal, the use index of the container processor is easy to increase, the performance of other service is affected, when the use index of the container processor exceeds a certain threshold value, the container is even restarted, and the current system monitoring mechanism can warn of the abnormality after the use rate of the processor exceeds the limit.
However, when the staff receives the warning information, the container is restarted, especially when the time is equal to the day of the bill day, the transaction amount is greatly increased, and if the abnormality is not found in time and the capacity is expanded in time, the service performance of the system is greatly reduced, and the operation efficiency of the business handling system is reduced.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a container resource monitoring method, apparatus, computer device, computer readable storage medium, and computer program product that can improve the operating efficiency of a business transaction system.
In a first aspect, the present application provides a method for monitoring a container resource. The method comprises the following steps:
acquiring container resource use information of a business object in a business handling system; the container resource use information comprises historical use information and real-time use information;
training the long-period memory network model according to the historical use information to obtain a first trend prediction model, and training the long-period memory network model according to the real-time use information to obtain a second trend prediction model; the first trend prediction model is used for outputting the whole prediction information of the current day; the holiday forecast information comprises forecast information for the holiday container resource usage trend; the second trend prediction model is used for outputting target period prediction information; the target period prediction information comprises container resource use trend prediction information of a target period; the target time period comprises a time period after the current time of day;
under the condition that the target period prediction information meets a preset alarm threshold value, determining a peak period initiation type of a service peak in the target period in the current day according to the current day overall prediction information;
And outputting a processing measure prompt message matched with the peak time initiation type.
In one embodiment, the determining, according to the global prediction information of the current day, a peak period initiation type of a peak of the service occurring in the target period in the current day includes:
under the condition that the date of the current day belongs to the estimated peak date and the container resource use trend before the current moment of the current day accords with the predicted trend in the overall prediction information of the current day, determining that the peak period initiation type of the service peak in the target period in the current day is a normal peak type;
wherein the estimated peak date includes a date of occurrence of a peak estimated based on the historical usage information.
In one embodiment, the determining, according to the global prediction information of the current day, a peak period initiation type of a peak of the service occurring in the target period in the current day includes:
under the condition that the date of the current day does not belong to the estimated peak date and the resource utilization rate is continuously increased, determining that the peak-time initiation type of the service peak in the target period in the current day is an abnormal peak type;
or determining that the type of the peak initiation of the service peak in the target time period in the current day is an abnormal peak type under the condition that the container resource use trend before the current time of the current day does not accord with the prediction trend in the overall prediction information of the current day and the resource use rate continuously rises.
In one embodiment, the peak time initiation type includes a normal peak time type, and the outputting the processing measure prompt information matched with the peak time initiation type includes:
under the condition that the peak period initiation type is the normal peak type, acquiring a capacity expansion processing strategy aiming at a service peak;
executing capacity expansion operation corresponding to the capacity expansion processing strategy to obtain a strategy implementation result;
and acquiring container log information, and outputting the policy implementation result and the container log information as the processing measure prompt information.
In one embodiment, the peak time initiation type includes an abnormal peak time type, and the outputting the processing measure prompt information matched with the peak time initiation type includes:
acquiring container log information under the condition that the peak period initiation type is the abnormal peak type;
and outputting the container log information as the processing measure prompt information.
In one embodiment, after the step of obtaining the container resource usage information of the business object in the business transaction system, the method further comprises:
determining an optimization objective for the container resource usage information prior to optimization;
Determining parameters of a sea squirt swarm algorithm according to the optimization target and the data characteristics of the container resource use information before optimization;
initializing according to the parameters of the goblet sea squirt swarm algorithm to generate an initial population;
updating the positions of leader individuals and follower individuals in the initial population to obtain a target population;
calculating the fitness of individuals in the target population;
and under the condition that the fitness meets the iteration termination condition, taking the position data represented in the target population as updated container resource use information.
In one embodiment, the updating the positions of the leader individuals and the follower individuals in the initial population to obtain the target population includes:
determining a target population individual updating strategy according to the iteration times; the population individual updating strategy comprises a leader fusion variable strategy and an optimal individual leading movement strategy;
updating the positions of leader individuals in the initial population according to the target population individual updating strategy;
updating the positions of follower individuals in the initial population according to the updated positions of the leader individuals;
and obtaining the target population according to the updated positions of the leader individuals and the updated follower individuals.
In a second aspect, the present application also provides a container resource monitoring device. The device comprises:
the container resource use information acquisition module is used for acquiring the container resource use information of the business object in the business handling system; the container resource use information comprises historical use information and real-time use information;
the model training module is used for training the long-period memory network model according to the historical use information to obtain a first trend prediction model, and training the long-period memory network model according to the real-time use information to obtain a second trend prediction model; the first trend prediction model is used for outputting the whole prediction information of the current day; the holiday forecast information comprises forecast information for the holiday container resource usage trend; the second trend prediction model is used for outputting target period prediction information; the target period prediction information comprises container resource use trend prediction information of a target period; the target time period comprises a time period after the current time of day;
the peak time initiation type determining module is used for determining the peak time initiation type of the service peak in the target time according to the whole current day prediction information under the condition that the target time prediction information meets a preset alarm threshold;
And the processing measure prompt information output module is used for outputting the processing measure prompt information matched with the peak time initiation type.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor which when executing the computer program performs the steps of:
acquiring container resource use information of a business object in a business handling system; the container resource use information comprises historical use information and real-time use information;
training the long-period memory network model according to the historical use information to obtain a first trend prediction model, and training the long-period memory network model according to the real-time use information to obtain a second trend prediction model; the first trend prediction model is used for outputting the whole prediction information of the current day; the holiday forecast information comprises forecast information for the holiday container resource usage trend; the second trend prediction model is used for outputting target period prediction information; the target period prediction information comprises container resource use trend prediction information of a target period; the target time period comprises a time period after the current time of day;
Under the condition that the target period prediction information meets a preset alarm threshold value, determining a peak period initiation type of a service peak in the target period in the current day according to the current day overall prediction information;
and outputting a processing measure prompt message matched with the peak time initiation type.
In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:
acquiring container resource use information of a business object in a business handling system; the container resource use information comprises historical use information and real-time use information;
training the long-period memory network model according to the historical use information to obtain a first trend prediction model, and training the long-period memory network model according to the real-time use information to obtain a second trend prediction model; the first trend prediction model is used for outputting the whole prediction information of the current day; the holiday forecast information comprises forecast information for the holiday container resource usage trend; the second trend prediction model is used for outputting target period prediction information; the target period prediction information comprises container resource use trend prediction information of a target period; the target time period comprises a time period after the current time of day;
Under the condition that the target period prediction information meets a preset alarm threshold value, determining a peak period initiation type of a service peak in the target period in the current day according to the current day overall prediction information;
and outputting a processing measure prompt message matched with the peak time initiation type.
In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the steps of:
acquiring container resource use information of a business object in a business handling system; the container resource use information comprises historical use information and real-time use information;
training the long-period memory network model according to the historical use information to obtain a first trend prediction model, and training the long-period memory network model according to the real-time use information to obtain a second trend prediction model; the first trend prediction model is used for outputting the whole prediction information of the current day; the holiday forecast information comprises forecast information for the holiday container resource usage trend; the second trend prediction model is used for outputting target period prediction information; the target period prediction information comprises container resource use trend prediction information of a target period; the target time period comprises a time period after the current time of day;
Under the condition that the target period prediction information meets a preset alarm threshold value, determining a peak period initiation type of a service peak in the target period in the current day according to the current day overall prediction information;
and outputting a processing measure prompt message matched with the peak time initiation type.
According to the container resource monitoring method, the device, the computer equipment, the storage medium and the computer program product, firstly, the container resource use information of a business object in a business handling system is acquired, then the long-short-period memory network model is trained according to the historical use information to obtain a first trend prediction model, the long-short-period memory network model is trained according to the real-time use information to obtain a second trend prediction model, further, under the condition that the target period prediction information meets a preset alarm threshold value, the peak time initiation type of the business peak in the target period in the current day is determined according to the current day integral prediction information, and finally, the processing measure prompt information matched with the peak time initiation type is output, so that the timeliness problem that the container resource use rate is exposed after being affected is solved, CPU use rate trend analysis is carried out by combining with the time periodicity and the planeness of a credit card system, the container resource use rate of the business handling system is automatically and timely emergency according to the intelligent matching decision scheme of different trend conditions, the manual emergency analysis time is reduced, the manual emergency risk is promoted, and further, the business handling efficiency of the business object is improved.
Drawings
FIG. 1 is a diagram of an application environment for a container resource monitoring method in one embodiment;
FIG. 2 is a flow diagram of a method of monitoring container resources in one embodiment;
FIG. 3 is a flow chart of a method for monitoring container resources according to another embodiment;
FIG. 4 is a flow diagram of data optimization of a container resource monitoring method in one embodiment;
FIG. 5 is a schematic diagram of a method of monitoring container resources in another embodiment;
FIG. 6 is a block diagram of a container resource monitoring device in one embodiment;
FIG. 7 is an internal block diagram of a computer device in one embodiment;
fig. 8 is an internal structural view of a computer device in another embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
It should be noted that, the user information (including, but not limited to, user equipment information, user personal information, etc.) and the data (including, but not limited to, data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data are required to comply with the related laws and regulations and standards of the related countries and regions.
It should be noted that the container resource monitoring method, apparatus, computer device, storage medium and computer program product disclosed in the present application may be applied to the field of financial technology, and may also be applied to any field other than the field of financial technology.
The container resource monitoring method provided by the embodiment of the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network.
The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, and portable wearable devices, where the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.
In one embodiment, as shown in fig. 2, a container resource monitoring method is provided, and the method is applied to the terminal 102 in fig. 1 for illustration, and includes the following steps:
s201, obtaining the container resource use information of the business object in the business handling system.
The service object comprises, but is not limited to, a financial institution such as a bank for handling related services, the service handling system is a software system for handling services for users, the container resource use information is information reflecting the use condition of the container resource in the service handling system, and concretely, the container resource use information comprises historical use information and real-time use information, wherein the historical use information reflects the historical use condition of the container resource, and the real-time use information reflects the real-time use condition of the container resource.
Illustratively, the historical usage information may be near day data, e.g., with 7 days as a period, and daily container information is collected during the period; real-time usage may be near real-time data, such as container data per moment of time during the day every half hour.
S202, training the long-period memory network model according to the historical use information to obtain a first trend prediction model, and training the long-period memory network model according to the real-time use information to obtain a second trend prediction model.
The first trend prediction model is used for outputting day overall prediction information, and the day overall prediction information comprises prediction information aiming at the overall use trend of the day container resources. Illustratively, the first trend prediction model is an LSTM model trained from the recent day data for predicting the total trend of use on the current day.
The second trend prediction model is used for outputting target period prediction information, and the target period prediction information comprises container resource use trend prediction information of a target period; the target period includes a period after the current time of day. The second trend prediction model is an LSTM model trained according to near real-time data, and is used for predicting a trend of a next period, for example, predicting a trend change condition of one hour after the current moment.
And S203, under the condition that the target period prediction information meets the preset alarm threshold, determining the peak period initiation type of the service peak in the target period in the current day according to the overall prediction information of the current day.
The peak period initiation types comprise a normal peak type and an abnormal peak type, wherein the normal peak type refers to a certain specific time node with larger flow, increased transaction amount, heavy container load and cpu high stroke initiated in the service peak period, such as deduction day, activity day and the like; the abnormal peak type refers to abnormal service causing cpu to continuously rise until overrun.
S204, outputting the processing measure prompt information matched with the peak time initiation type.
The processing measure prompt information is used for guiding system maintenance personnel to conduct emergency processing.
For the trend of normal business peak, the temporary capacity expansion treatment is implemented in linkage paas emergency, the performance capacity of the container is improved to support the flow in peak, the influence on user experience is avoided, meanwhile, the container log and the implementation result of capacity expansion operation are pulled and sent to system maintenance personnel, and whether permanent capacity expansion is needed is prompted and analyzed; and aiming at the abnormal rising trend, automatically pulling log file backups of application logs, health check logs, java files, NMON files and the like in real time, and notifying system maintenance personnel of timely analysis and processing.
According to the container resource monitoring method, firstly, the container resource use information of a business object in a business handling system is obtained, then, a long-period memory network model is trained according to historical use information to obtain a first trend prediction model, and the long-period memory network model is trained according to real-time use information to obtain a second trend prediction model, further, under the condition that target period prediction information meets a preset alarm threshold value, the peak time initiation type of a business peak in the target period in the current day is determined according to the whole prediction information of the current day, and finally, processing measure prompt information matched with the peak time initiation type is output, so that the timeliness problem that the container resource use efficiency is exposed after the abnormality occurs and is influenced is solved, CPU use efficiency trend analysis is conducted by combining with the timeliness and the timeliness of a credit card system, a decision scheme is automatically and timely emergency is matched according to different trend conditions, manual analysis time is shortened, manual emergency risk is reduced, the container resource use efficiency of the business handling system is improved, and business handling experience of a user on the business object is further improved.
In one embodiment, determining a peak time initiation type for a peak of traffic occurring in a target time period on the current day based on the current day overall prediction information comprises: under the condition that the date of the current day belongs to the estimated peak date and the container resource use trend before the current time of the current day accords with the prediction trend in the overall prediction information of the current day, determining that the peak period initiation type of the service peak in the target period in the current day is a normal peak type;
wherein, the estimated peak date comprises the estimated peak date according to the historical use information; the normal peak type refers to a certain specific time when node flow is large, transaction amount is increased, container load is increased, and cpu is raised in the peak period of business, such as deduction day, activity day and the like.
When the trend at the next moment reaches the alarm threshold, whether the current day matches the special time sequence of the credit card is judged, and whether the trend of the current day container use situation accords with the predicted current day use total trend is judged, if so, the current day container use situation is judged to be the trend of the normal business peak period.
In this embodiment, under the condition that the date of the current day belongs to the estimated peak date and the container resource usage trend before the current time of the current day accords with the prediction trend in the overall prediction information of the current day, the peak time initiation type of the service peak in the target time in the current day is determined to be the normal peak type, the judgment basis of the normal peak type is described, the influence of the estimated peak date on the container resource is considered, the influence of the prediction condition of the current day trend is considered, the two are taken as the division basis of the usage peak time, the peak time is determined to be the normal condition when the two conditions are simultaneously satisfied, the accuracy of judging the initiation reason of the peak time is improved, and the accuracy of emergency measure adoption is further improved.
In one embodiment, determining a peak time initiation type for a peak of traffic occurring in a target time period on the current day based on the current day overall prediction information comprises: under the condition that the date of the current day does not belong to the estimated peak date and the resource utilization rate is continuously increased, determining that the peak-time initiation type of the service peak in the target period in the current day is an abnormal peak type; or under the condition that the container resource use trend before the current moment of the day does not accord with the prediction trend in the overall prediction information of the day and the resource use rate continuously rises, determining that the peak-time initiation type of the service peak in the target period in the day is an abnormal peak type.
Wherein, the abnormal peak type refers to abnormal service causing cpu to continuously rise until overrun.
Illustratively, it is determined whether the current day matches the credit card special time series, and whether the trend of the current day container usage matches the predicted current day usage total trend, if one of the trends does not match the condition, and if the resource usage continues to rise, the decision is an abnormally rising trend.
In this embodiment, when the date of the current day does not belong to the estimated peak date or the container resource usage trend before the current time of the current day does not conform to the predicted trend in the overall predicted information of the current day, and the resource usage rate is continuously increased, the peak period initiation type of the service peak occurring in the target period of the current day is determined to be an abnormal peak type, the judgment basis of the abnormal peak type is described, the influence of the estimated peak date on the container resource is considered, the influence of the predicted condition of the current day is considered, and the two are taken as the division basis of the usage peak period, and when the two conditions are not satisfied at the same time and the resource usage rate is continuously increased, the peak period is judged to be a normal condition, the accuracy of judging the cause of the peak period is improved, and the accuracy of emergency measure adoption is further improved.
In one embodiment, the peak time initiation type includes a normal peak type, and outputting a processing measure hint information that matches the peak time initiation type includes: under the condition that the peak period initiation type is a normal peak type, acquiring a capacity expansion processing strategy aiming at a service peak; executing capacity expansion operation corresponding to the capacity expansion processing strategy to obtain a strategy implementation result; and acquiring container log information, and outputting a policy implementation result and the container log information as processing measure prompt information.
For the trend of normal business peak, the temporary capacity expansion treatment is implemented in linkage paas emergency, the performance capacity of the container is improved to support the flow in peak, the user experience is prevented from being influenced, meanwhile, the container log is pulled, the implementation result of capacity expansion operation is sent to system maintenance personnel, and whether permanent capacity expansion is needed is prompted to analyze.
In this embodiment, a capacity expansion processing policy for a service peak is obtained first according to a situation of a normal peak type, then a capacity expansion operation corresponding to the capacity expansion processing policy is executed, a policy implementation result is obtained, further container log information is obtained, the policy implementation result and the container log information are output as processing measure prompt information, triggering of an emergency capacity expansion operation under the situation of the normal peak type is completed, and the implementation result and the log information are timely sent to a system maintainer for further data analysis by the system maintainer, so that scientificity of emergency processing is improved, and efficiency of container resource utilization is further ensured.
In one embodiment, the peak time initiation type includes an abnormal peak time type, and outputting a processing measure hint information that matches the peak time initiation type, including: under the condition that the peak period initiation type is an abnormal peak type, acquiring container log information; and outputting the container log information as processing measure prompt information.
For example, for abnormal rising trend, log file backups such as application logs, health check logs, java files, NMON files and the like are automatically pulled in real time, and system maintenance personnel are notified of timely analysis and processing.
In this embodiment, the container log information is firstly obtained for the abnormal peak type condition, then the container log information is output as the processing measure prompt information, and the corresponding log information is timely sent to the system maintainer for further data analysis by the system maintainer, so that the processing efficiency of emergency processing for the abnormal condition is improved, and the efficiency of container resource utilization is further ensured.
In one embodiment, after the step of obtaining container resource usage information of the business object in the business transaction system, the method further comprises: determining an optimization target for the pre-optimization container resource usage information; determining parameters of the goblet sea squirt swarm algorithm according to the optimization target and the data characteristics of the container resource use information before optimization; initializing according to parameters of the goblet sea squirt swarm algorithm to generate an initial population; updating the positions of leader individuals and follower individuals in the initial population to obtain a target population; calculating the fitness of individuals in the target population; and under the condition that the fitness meets the iteration termination condition, taking the position data represented in the target population as updated container resource use information.
The data accuracy is optimized by screening out the data which is used for a long time and is most effective for prediction, and extracting more accurate data for prediction. After the algorithm is realized, the current food position is output as the estimated position of the target, and the actual meaning of the data corresponding to the estimated position is the optimized optimal data.
Illustratively, each type of data within half an hour is input before the quasi-real-time data optimization, one piece of data is input for 15s, and 120 pieces of data are input for each type, for example, one piece of data can be: [ Container cpu information (C), container memory information (G), network received (kBs), network sent (kBs), file read/write Rate (kBs), number of threads used (C), file System storage usage (G), file System inode number ].
In this embodiment, an optimization target for the container resource usage information before optimization is first determined, then parameters of the goblet sea squirt swarm algorithm are determined according to the optimization target and the data characteristics of the container resource usage information before optimization, and then the parameters of the goblet sea squirt swarm algorithm are initialized to generate an initial population, positions of a leader individual and a follower individual in the initial population are updated to obtain a target population, then fitness of the individuals in the target population is calculated, finally, the position data represented in the target population are used as updated container resource usage information under the condition that the fitness meets iteration termination conditions, and data of the most suitable time step is selected after optimization, so that accuracy of subsequent trend analysis is improved.
In one embodiment, updating the positions of the leader individuals and follower individuals in the initial population to obtain the target population includes:
determining a target population individual updating strategy according to the iteration times; updating the positions of leader individuals in the initial population according to the target population individual updating strategy; updating the positions of follower individuals in the initial population according to the updated positions of the leader individuals; and obtaining a target population according to the updated positions of the leader individuals and the updated follower individuals.
The population individual updating strategy comprises a leader fusion variable strategy and an optimal individual leading movement strategy.
Illustratively, as shown in FIG. 4, when the current number of iterations reaches a maximum, the leader position is updated with the leader fusion variable policy, and when the current number of iterations does not meet the maximum, the leader position is updated with the optimal individual leader motion.
In this embodiment, the target population individual update strategy is determined according to the iteration number, then the positions of the leader individuals in the initial population are updated according to the target population individual update strategy, the positions of the follower individuals in the initial population are updated according to the updated positions of the leader individuals, finally the target population is obtained according to the updated positions of the leader individuals and the updated follower individuals, and the population is updated by combining the advantages of the two update strategies and utilizing different strategies under different situations, so that the optimization effect of the optimization algorithm on data is improved, further the accuracy of the subsequent trend analysis is improved, and the utilization efficiency of container resources is improved.
In another embodiment, as shown in fig. 3, a container resource monitoring method is provided, and the method is applied to the terminal 102 in fig. 1 for illustration, and includes the following steps:
s301, under the condition that the target period prediction information meets a preset alarm threshold, determining that the peak period initiation type of the service peak in the target period in the current day is a normal peak type under the condition that the date of the current day belongs to the estimated peak date and the container resource use trend before the current time of the current day accords with the prediction trend in the whole prediction information of the current day;
s302, under the condition that the date of the current day does not belong to the estimated peak date and the resource utilization rate is continuously increased, determining that the peak time initiation type of the service peak in the target time in the current day is an abnormal peak type;
s303, determining that the peak-period initiation type of the service peak in the target period in the current day is an abnormal peak type under the condition that the container resource use trend before the current time of the current day does not accord with the prediction trend in the overall prediction information of the current day and the resource use rate continuously rises.
S304, under the condition that the peak period initiation type is a normal peak type, acquiring a capacity expansion processing strategy aiming at a service peak;
S305, performing capacity expansion operation corresponding to the capacity expansion processing strategy to obtain a strategy implementation result;
s306, acquiring container log information, and outputting a policy implementation result and the container log information as processing measure prompt information.
S307, acquiring container log information under the condition that the peak time initiation type is an abnormal peak type;
s308, outputting the container log information as processing measure prompt information.
It should be noted that, the specific limitation of the above steps may be referred to the specific limitation of a container resource monitoring method, which is not described herein.
In some embodiments, the monitoring means for container resources can only find and process the abnormality when the abnormality occurs, and in the process, the data available for analysis is covered after restarting the container, which reduces the efficiency of problem analysis. Specifically, restarting the container results in the data available for analysis being covered, which in turn results in difficulty in timely troubleshooting the cause of the analysis, reducing the efficiency of the problem analysis.
Based on the method, the method for monitoring the container resources comprises a group intelligent optimization algorithm and an LSTM (long-short-term memory artificial neural network model), solves the timeliness problem that the abnormal utilization rate is exposed after the abnormal utilization rate is generated and influenced in the current production, can combine the time periodicity and the planeness of a credit card system to analyze the CPU utilization rate trend, and automatically and timely emergency according to intelligent matching decision schemes of different trend conditions, thereby effectively reducing the influence.
Specifically, in the optimization process, after global optimality optimization data of the ecteinascidia group algorithm are utilized, a long-term and short-term memory artificial neural network model is trained to respectively predict the total daily use trend and the next time trend, different emergency schemes are decided for different trends and are automatically processed, analysis cost of developers is reduced, emergency efficiency is improved, and time and labor are saved.
FIG. 4 provides a data optimization flow chart of a container resource monitoring method for ease of understanding by those skilled in the art; fig. 5 provides a schematic diagram of a method of monitoring container resources.
The container resource monitoring method is described in detail below in one particular embodiment with reference to fig. 4 and 5. It is to be understood that the following description is exemplary only and is not intended to limit the application to the details of construction and the arrangements of the components set forth herein.
The container resource monitoring device corresponding to the container resource monitoring method mainly comprises four modules, namely: the system comprises a data acquisition and optimization module, a trend analysis module, an intelligent decision module and an emergency processing module.
Data acquisition and optimization module
The data acquisition and optimization module optimizes acquired data by using a goblet sea squirt swarm algorithm, searches by taking global optimization as a center, ensures population convergence, optimizes data precision and selects optimal parameters; the trend analysis module uses the optimized data to train an LSTM model, predicts the total daily use trend according to the data of the same period of the past year and the daily use condition, timely discovers the performance influence possibly caused by insufficient container resources, predicts the trend of the next time according to the real-time data, and pre-warns and analyzes in advance so as to avoid the influence in a larger range; the intelligent decision-making module makes automatic intelligent decision-making for the predicted result, and proper decision-making is selected according to different trend directions; the emergency processing module performs automatic emergency processing according to a preset emergency scheme, reduces manual operation and smoothly handles risks. The individual modules of the device will be described in detail below.
The data acquisition module mainly acquires near-day data for predicting the total trend of the daily use and quasi-real-time data for predicting the trend of the next moment.
The quasi-real-time data mainly comprises container cpu information, container memory information, network throughput, file read-write speed, thread number, file system use information and the like, and is collected by taking each half hour as a time node; the file system usage information refers to file system storage usage (G), and the number of file system inodes.
The file system usage information refers to information for monitoring and counting the usage of the file system in the container, and comprises file system storage usage and file system inode numbers.
File system storage usage provides an indication of storage space usage of the file system in the container, including total capacity, used capacity, and available capacity. Such information may help an administrator to understand the storage status of the file system in the container for capacity planning and resource management. For example, when the storage space of the file system approaches or exceeds the upper limit of capacity, an administrator may take timely action, such as cleaning up garbage or expanding the file system, to avoid problems caused by insufficient storage space.
The file system inode number refers to a data structure in the file system for storing files and directory metadata. The file system inode number refers to the number of inodes used by the file system in the container. The size of the inode number limits the number of files and directories, so monitoring the file system inode number can help an administrator to know the number of files and directories in a container, as well as the use of inode resources. When the number of inodes of the file system approaches or reaches an upper limit, an administrator can take measures in time, such as cleaning up garbage files or adjusting the inode resource quota, so as to avoid problems caused by insufficient inode resources.
In summary, the file system usage information may provide the storage usage condition and inode number of the file system in the container, so as to help the administrator to perform capacity planning, resource management and performance optimization, so as to ensure normal operation and efficient utilization of the file system.
In addition, the periodic system time of the credit card, such as the activity date and the like sequence of the repayment date, is used as a key factor, such as the month-to-year growth rate of the repayment date, the activity date can collect the year-to-year growth rate of the transaction amount, and the prediction accuracy of the special date is improved.
For the data acquired above, the data acquisition module optimizes the data by using an algorithm, the implementation flow of which is shown in fig. 4, and the application can optimize the data by adopting a goblet sea squirt swarm algorithm.
The data accuracy is optimized by screening out the data which is used for a long time and is most effective for prediction, and extracting more accurate data for prediction. After the algorithm is realized, the current food position is output as the estimated position of the target, and the actual meaning of the data corresponding to the estimated position is the optimized optimal data.
Illustratively, each type of data within half an hour is input before the quasi-real-time data optimization, one piece of data is input for 15s, and 120 pieces of data are input for each type, for example, one piece of data can be: [ Container cpu information (C), container memory information (G), network received (kBs), network sent (kBs), file read/write Rate (kBs), number of threads used (C), file System storage usage (G), file System inode number ].
The data optimization means provided by the application combines the existing container cpu information to optimize the rest various data, and selects the data with the most proper time step after optimization, thereby improving the accuracy of the follow-up trend analysis.
(II) trend analysis module
The optimized data will be used to train LSTM model for prediction, and the LSTM unit chain structure diagram is shown in FIG. 5.
First define cell state C t-1 Which features will be used for calculation, f t As a forget gate containing a sigmoid layer, for an input x t And h t-1 Output an intermediate of [0,1 ]]The value of the interval, 0 indicates all deletions, and 1 indicates all reservations.
f t =σ(W f ·[h t-1 ,x t ]+b f )
See again how the cell state adds new messages, i t Is an input gate, which is formed by inputting data x t And hidden node h t-1 Calculated via sigmoid activation function to be between [0,1 ]]The value of the interval is then given by a tanh layer x t And h t-1 Generating a cell state update valuei t Can control +.>Which features of (a) are used to update C t
i t =σ(W i ·[h t-1 ,x t ]+b i )
The cell state can then be updated with the forget gate and the input gate, e.g., as shown in the following equation, f t ×C t-1 Information indicating that deletion is desired,representing the newly added information.
Finally, calculating the output h of the hidden node t From the output gate o t And cell state C t Obtaining an output gate for determining LSTM output content, o t Calculated from sigmoid activation function, C is obtained by using tanh layer t Push to [ -1,1]And multiplies it by the output of the sigmoid layer to output only the desired portion.
o t =σ(W o [h t-1 ,x t ]+b o )
h t =o t *tanh(C t )
The trend analysis module trains out a model for effectively predicting the total daily use trend and the next moment use trend through the processes by using different data sets, and outputs a future one-hour container resource predicted value according to container data in each half hour of the day; and outputting the general container resource prediction trend of the current day according to the container information and the same-ratio data of each moment of the day of the seven days in the period.
(III) Intelligent decision module
CPU flush is generally responsible for two reasons, one is that abnormal service causes CPU to continue to rise until overrun; the other is that some specific time nodes such as deduction days, activity days and the like have larger flow time, the transaction amount is increased, the container load is increased, and the cpu is raised in the business peak period.
The intelligent decision module can preset an alarm threshold and a decision direction, and intelligently match decision schemes according to different trend conditions. When the trend at the next moment reaches the alarm threshold, judging whether the current day is matched with the special time sequence of the credit card, and judging whether the trend of the current day container use condition is consistent with the predicted current day use total trend, if so, determining that the current day container use condition is a normal business peak trend, and if not, determining that the current day container use condition is an abnormal rising trend. The above decision will be passed to the emergency module to immediately perform the emergency operation.
(IV) Emergency treatment Module
According to the judgment of the decision-making module, carrying out temporary capacity expansion treatment on the normal business peak trend in a linked paas emergency mode, improving the performance capacity of the container to support the peak flow, avoiding affecting the user experience, simultaneously pulling the container log and the capacity expansion operation implementation result to be sent to system maintenance personnel, and prompting whether permanent capacity expansion is needed or not for analysis;
Whether permanent capacity expansion is needed to be analyzed is not needed by the module, and the module is comprehensively analyzed by system maintenance personnel according to the conditions of system service requirements, functions and the like, and only needs to implement temporary capacity expansion for emergency and prompt.
And aiming at the abnormal rising trend, automatically pulling log file backups of application logs, health check logs, java files, NMON files and the like in real time, and notifying system maintenance personnel of timely analysis and processing.
It should be noted that, the log data is mainly used for analyzing and locating the abnormal problem by the system maintainer, and making different emergency schemes according to different analysis results, and compared with the log data which is obtained after restarting the container due to the abnormality, the log data is more perfect. The normal business peak scene does not need to analyze the problem by using the log data as the abnormal scene, and the log data is only used for assisting in observation.
According to the container resource monitoring method, the use condition of the container CPU is predicted by deep learning, the problem that the original CPU can only be perceived after abnormality occurs is solved, various log files are pulled in real time, the problem that the original container is restarted and the log information available for analysis is less is also optimized, and the analysis difficulty of developers is reduced.
In addition, the container resource monitoring method provided by the application not only can predict the abnormality in advance and improve the aging, but also can strive for precious time for abnormality analysis, and greatly reduces the service influence possibly brought in the future. For non-service abnormal conditions such as business peak period, automatic emergency treatment can be performed in time, so that manual analysis time and manual emergency risk are reduced, and influence on user experience is directly avoided.
It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, the embodiment of the application also provides a container resource monitoring device for realizing the above-mentioned container resource monitoring method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiments of the container resource monitoring device or devices provided below may be referred to the limitation of the container resource monitoring method hereinabove, and will not be described herein.
In one embodiment, as shown in fig. 6, there is provided a container resource monitoring apparatus comprising: a container resource usage information acquisition module 601, a model training module 602, a peak period initiation type determination module 603, and a processing measure prompt information output module 604, wherein:
a container resource usage information obtaining module 601, configured to obtain container resource usage information of a service object in a service handling system; the container resource usage information includes historical usage information and real-time usage information;
the model training module 602 is configured to train the long-short-term memory network model according to the historical usage information to obtain a first trend prediction model, and train the long-short-term memory network model according to the real-time usage information to obtain a second trend prediction model; the first trend prediction model is used for outputting the whole prediction information of the current day; the holiday forecast information includes forecast information for a holiday container resource usage trend; the second trend prediction model is used for outputting target period prediction information; the target period prediction information includes container resource usage trend prediction information of the target period; the target period includes a period after the current time of day;
The peak initiation type determining module 603 is configured to determine, according to the overall prediction information of the current day, a peak initiation type of a service peak occurring in the target period in the current day, if the target period prediction information meets a preset alarm threshold;
and the processing measure prompt information output module 604 is used for outputting the processing measure prompt information matched with the peak time initiation type.
In one embodiment, the rush hour initiation type determination module is further for: under the condition that the date of the current day belongs to the estimated peak date and the container resource use trend before the current time of the current day accords with the prediction trend in the overall prediction information of the current day, determining that the peak period initiation type of the service peak in the target period in the current day is a normal peak type; wherein, the estimated peak date comprises the estimated peak date according to the historical use information.
In one embodiment, the rush hour initiation type determination module is further for: under the condition that the date of the current day does not belong to the estimated peak date and the resource utilization rate is continuously increased, determining that the peak-time initiation type of the service peak in the target period in the current day is an abnormal peak type; or under the condition that the container resource use trend before the current moment of the day does not accord with the prediction trend in the overall prediction information of the day and the resource use rate continuously rises, determining that the peak-time initiation type of the service peak in the target period in the day is an abnormal peak type.
In one embodiment, the processing measure prompt information output module is further configured to: under the condition that the peak period initiation type is a normal peak type, acquiring a capacity expansion processing strategy aiming at a service peak; executing capacity expansion operation corresponding to the capacity expansion processing strategy to obtain a strategy implementation result; and acquiring container log information, and outputting a policy implementation result and the container log information as processing measure prompt information.
In one embodiment, the processing measure prompt information output module is further configured to: under the condition that the peak period initiation type is an abnormal peak type, acquiring container log information; and outputting the container log information as processing measure prompt information.
In one embodiment, the apparatus is further to: determining an optimization target for the pre-optimization container resource usage information; determining parameters of the goblet sea squirt swarm algorithm according to the optimization target and the data characteristics of the container resource use information before optimization; initializing according to parameters of the goblet sea squirt swarm algorithm to generate an initial population; updating the positions of leader individuals and follower individuals in the initial population to obtain a target population; calculating the fitness of individuals in the target population; and under the condition that the fitness meets the iteration termination condition, taking the position data represented in the target population as updated container resource use information.
In one embodiment, the apparatus is further to: determining a target population individual updating strategy according to the iteration times; the population individual updating strategy comprises a leader fusion variable strategy and an optimal individual leading movement strategy; updating the positions of leader individuals in the initial population according to the target population individual updating strategy; updating the positions of follower individuals in the initial population according to the updated positions of the leader individuals; and obtaining a target population according to the updated positions of the leader individuals and the updated follower individuals.
The various modules in the container resource monitoring device described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is for storing data. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a container resource monitoring method.
In one embodiment, a computer device is provided, which may be a terminal, and the internal structure thereof may be as shown in fig. 8. The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input means. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a container resource monitoring method. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structures shown in fig. 7 and 8 are block diagrams of only some of the structures associated with the present application and are not intended to limit the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device includes a memory having a computer program stored therein and a processor that when executing the computer program performs the steps of the method embodiments described above.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, carries out the steps of the method embodiments described above.
In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the method embodiments described above.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims (11)

1. A method of monitoring a container resource, the method comprising:
acquiring container resource use information of a business object in a business handling system; the container resource use information comprises historical use information and real-time use information;
training the long-period memory network model according to the historical use information to obtain a first trend prediction model, and training the long-period memory network model according to the real-time use information to obtain a second trend prediction model; the first trend prediction model is used for outputting the whole prediction information of the current day; the holiday forecast information comprises forecast information for the holiday container resource usage trend; the second trend prediction model is used for outputting target period prediction information; the target period prediction information comprises container resource use trend prediction information of a target period; the target time period comprises a time period after the current time of day;
Under the condition that the target period prediction information meets a preset alarm threshold value, determining a peak period initiation type of a service peak in the target period in the current day according to the current day overall prediction information;
and outputting a processing measure prompt message matched with the peak time initiation type.
2. The method of claim 1, wherein determining a peak initiation type for a peak of traffic occurring in the target period of the day based on the day overall prediction information comprises:
under the condition that the date of the current day belongs to the estimated peak date and the container resource use trend before the current moment of the current day accords with the predicted trend in the overall prediction information of the current day, determining that the peak period initiation type of the service peak in the target period in the current day is a normal peak type;
wherein the estimated peak date includes a date of occurrence of a peak estimated based on the historical usage information.
3. The method according to claim 2, wherein determining a peak time initiation type of a peak of traffic occurring in the target time period on the current day based on the current day overall prediction information comprises:
Under the condition that the date of the current day does not belong to the estimated peak date and the resource utilization rate is continuously increased, determining that the peak-time initiation type of the service peak in the target period in the current day is an abnormal peak type;
or determining that the type of the peak initiation of the service peak in the target time period in the current day is an abnormal peak type under the condition that the container resource use trend before the current time of the current day does not accord with the prediction trend in the overall prediction information of the current day and the resource use rate continuously rises.
4. The method of claim 1, wherein the rush hour initiation type comprises a normal rush hour type, and wherein outputting a treatment action prompt matching the rush hour initiation type comprises:
under the condition that the peak period initiation type is the normal peak type, acquiring a capacity expansion processing strategy aiming at a service peak;
executing capacity expansion operation corresponding to the capacity expansion processing strategy to obtain a strategy implementation result;
and acquiring container log information, and outputting the policy implementation result and the container log information as the processing measure prompt information.
5. The method of claim 4, wherein the rush hour initiation type comprises an abnormal rush hour type, and wherein outputting a treatment action prompt matching the rush hour initiation type comprises:
Acquiring container log information under the condition that the peak period initiation type is the abnormal peak type;
and outputting the container log information as the processing measure prompt information.
6. The method of claim 1, wherein after the step of obtaining container resource usage information of the business object in the business transaction system, the method further comprises:
determining an optimization objective for the container resource usage information prior to optimization;
determining parameters of a sea squirt swarm algorithm according to the optimization target and the data characteristics of the container resource use information before optimization;
initializing according to the parameters of the goblet sea squirt swarm algorithm to generate an initial population;
updating the positions of leader individuals and follower individuals in the initial population to obtain a target population;
calculating the fitness of individuals in the target population;
and under the condition that the fitness meets the iteration termination condition, taking the position data represented in the target population as updated container resource use information.
7. The method of claim 6, wherein updating the locations of the leader individuals and follower individuals in the initial population to obtain a target population comprises:
Determining a target population individual updating strategy according to the iteration times; the population individual updating strategy comprises a leader fusion variable strategy and an optimal individual leading movement strategy;
updating the positions of leader individuals in the initial population according to the target population individual updating strategy;
updating the positions of follower individuals in the initial population according to the updated positions of the leader individuals;
and obtaining the target population according to the updated positions of the leader individuals and the updated follower individuals.
8. A container resource monitoring device, the device comprising:
the container resource use information acquisition module is used for acquiring the container resource use information of the business object in the business handling system; the container resource use information comprises historical use information and real-time use information;
the model training module is used for training the long-period memory network model according to the historical use information to obtain a first trend prediction model, and training the long-period memory network model according to the real-time use information to obtain a second trend prediction model; the first trend prediction model is used for outputting the whole prediction information of the current day; the holiday forecast information comprises forecast information for the holiday container resource usage trend; the second trend prediction model is used for outputting target period prediction information; the target period prediction information comprises container resource use trend prediction information of a target period; the target time period comprises a time period after the current time of day;
The peak time initiation type determining module is used for determining the peak time initiation type of the service peak in the target time according to the whole current day prediction information under the condition that the target time prediction information meets a preset alarm threshold;
and the processing measure prompt information output module is used for outputting the processing measure prompt information matched with the peak time initiation type.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
11. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
CN202311489628.8A 2023-11-09 2023-11-09 Container resource monitoring method, device, computer equipment and storage medium Pending CN117472511A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311489628.8A CN117472511A (en) 2023-11-09 2023-11-09 Container resource monitoring method, device, computer equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311489628.8A CN117472511A (en) 2023-11-09 2023-11-09 Container resource monitoring method, device, computer equipment and storage medium

Publications (1)

Publication Number Publication Date
CN117472511A true CN117472511A (en) 2024-01-30

Family

ID=89623576

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311489628.8A Pending CN117472511A (en) 2023-11-09 2023-11-09 Container resource monitoring method, device, computer equipment and storage medium

Country Status (1)

Country Link
CN (1) CN117472511A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117972069A (en) * 2024-04-01 2024-05-03 南京信人智能科技有限公司 Method for carrying out active dialogue and knowledge base vector search based on artificial intelligence

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117972069A (en) * 2024-04-01 2024-05-03 南京信人智能科技有限公司 Method for carrying out active dialogue and knowledge base vector search based on artificial intelligence
CN117972069B (en) * 2024-04-01 2024-05-28 南京信人智能科技有限公司 Method for carrying out active dialogue and knowledge base vector search based on artificial intelligence

Similar Documents

Publication Publication Date Title
US20210264272A1 (en) Training method and system of neural network model and prediction method and system
NO20230419A1 (en) Computer-based systems, computing components and computing objects configured to implement dynamic outlier bias reduction in machine learning models
CN111611478B (en) Information recommendation method and device and electronic equipment
US20180096253A1 (en) Rare event forecasting system and method
Esquivel et al. Spatio-temporal prediction of Baltimore crime events using CLSTM neural networks
CN110705719A (en) Method and apparatus for performing automatic machine learning
CN116611546B (en) Knowledge-graph-based landslide prediction method and system for target research area
CN111582341B (en) User abnormal operation prediction method and device
CN111340240A (en) Method and device for realizing automatic machine learning
WO2023207411A1 (en) Traffic determination method and apparatus based on spatio-temporal data, and device and medium
CN114118507A (en) Risk assessment early warning method and device based on multi-dimensional information fusion
CN115564152A (en) Carbon emission prediction method and device based on STIRPAT model
CN117472511A (en) Container resource monitoring method, device, computer equipment and storage medium
CN112862014A (en) Client credit early warning method and device
CN115883392B (en) Data perception method and device of computing power network, electronic equipment and storage medium
CN117389780A (en) Converter station fault analysis method, device, computer equipment and storage medium
Keating et al. Using decision analysis to determine the feasibility of a conservation translocation
CN111737319B (en) User cluster prediction method, device, computer equipment and storage medium
CN115600818A (en) Multi-dimensional scoring method and device, electronic equipment and storage medium
CN112926803A (en) Client deposit loss condition prediction method and device based on LSTM network
CN118134054B (en) Model training method and space-time data prediction method
CN118279113B (en) Digital intelligence social public service management method and system based on large model
CN113162780B (en) Real-time network congestion analysis method, device, computer equipment and storage medium
Plessis et al. Unsupervised multi scale anomaly detection in streams of events
CN117057736A (en) Digital twin city management platform based on intelligent brain

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