Business monitoring methods, devices, electronic equipment and computer-readable storage media

Abstract

This invention provides a business monitoring method, apparatus, electronic device, and computer-readable storage medium. The method includes: acquiring a first data point sequence of a business entity, wherein the first data point sequence includes multiple data points obtained from collecting data on the business entity's business at multiple time points; taking the data point to be monitored among the multiple data points as a starting point, extracting multiple data points from the first data point sequence whose collection time is before the starting point and falls within a first predetermined time period, and constructing a second data point sequence from the data point to be monitored and the extracted multiple data points; performing boundary division processing on the second data point sequence to obtain an upper limit and a lower limit of business data for the data point to be monitored; comparing the business data of the data point to be monitored with the upper limit and lower limit of business data to obtain the fluctuation trend of the data point to be monitored, which serves as the business monitoring result for the business entity. Through this invention, accurate and timely monitoring of business operations is possible.

Description

Technical Field

[0001] This invention relates to big data and monitoring technology, and more particularly to a business monitoring method, apparatus, electronic device, and computer-readable storage medium. Background Technology

[0002] With the proliferation of various internet-based services, the production environment supporting these services includes a large number of servers, which generate a significant amount of business data during the service delivery process.

[0003] Business monitoring refers to the collection and detection of serialized data generated in the production environment of a business entity at a series of time points, and the issuance of alerts when necessary, so that the business operations team can respond in a timely manner to ensure the stable and orderly operation of the business.

[0004] In the solutions provided by related technologies, due to the limitations of monitoring granularity, it is impossible to monitor abnormal business states in a timely manner, that is, the accuracy of business monitoring is low. Summary of the Invention

[0005] This invention provides a business monitoring method, apparatus, electronic device, and computer-readable storage medium, which can accurately and timely monitor business operations.

[0006] The technical solution of this invention is implemented as follows:

[0007] This invention provides a business monitoring method, including:

[0008] Obtain the first data point sequence of the business party, wherein the first data point sequence includes multiple data points obtained by collecting data on the business party's business at multiple time points;

[0009] Starting from the data point to be monitored among the multiple data points, multiple data points that were collected before the starting point and are within a first set time period are extracted from the first data point sequence. The data point to be monitored and the extracted multiple data points constitute a second data point sequence.

[0010] The second data point sequence is subjected to boundary division processing to obtain the upper limit and lower limit of the business data of the data points to be monitored.

[0011] The business data of the data point to be monitored is compared with the upper limit and lower limit of the business data to obtain the fluctuation trend of the data point to be monitored, which serves as the business monitoring result for the business party.

[0012] This invention provides a business monitoring device, comprising:

[0013] The acquisition module is used to acquire a first data point sequence of the business party, wherein the first data point sequence includes multiple data points obtained by collecting data on the business party's business at multiple time points.

[0014] The interception module is used to take the data point to be monitored among the multiple data points as the starting point, and intercept multiple data points in the first data point sequence whose collection time is before the starting point and are within a first set time period, and to form a second data point sequence by the data point to be monitored and the intercepted multiple data points.

[0015] The partitioning module is used to perform boundary partitioning processing on the second data point sequence to obtain the upper limit and lower limit of the business data of the data points to be monitored.

[0016] The comparison module is used to compare the business data of the data point to be monitored with the upper limit and the lower limit of the business data to obtain the fluctuation trend of the data point to be monitored, which serves as the business monitoring result for the business party.

[0017] This invention provides an electronic device, comprising:

[0018] Memory, used to store executable instructions;

[0019] The processor, when executing executable instructions stored in the memory, implements the service monitoring method provided in this embodiment of the invention.

[0020] This invention provides a computer-readable storage medium storing executable instructions, which, when executed by a processor, implement the business monitoring method provided in this invention.

[0021] The embodiments of the present invention have the following beneficial effects:

[0022] By taking the data point to be monitored as the starting point, extracting the second data point sequence from the first data point sequence, and performing boundary division processing on the second data point sequence to obtain the upper limit and lower limit of the business data of the data point to be monitored, the fluctuation trend of the data point to be monitored can be obtained, realizing business monitoring with the data point as the finest granularity, and enabling accurate and timely monitoring of the business. Attached Figure Description

[0023] Figure 1 This is an optional architecture diagram of the business monitoring system provided in an embodiment of the present invention;

[0024] Figure 2 This is an optional architecture diagram of a business monitoring system that integrates a blockchain network, provided in an embodiment of the present invention.

[0025] Figure 3This is a schematic diagram of an optional server architecture provided in an embodiment of the present invention;

[0026] Figure 4 This is a schematic diagram of an optional architecture of the business monitoring device provided in an embodiment of the present invention;

[0027] Figure 5A This is an optional flowchart illustrating the business monitoring method provided in this embodiment of the invention;

[0028] Figure 5B This is an optional flowchart illustrating the business monitoring method provided in this embodiment of the invention;

[0029] Figure 5C This is an optional flowchart illustrating the business monitoring method provided in this embodiment of the invention;

[0030] Figure 6A This is an optional schematic diagram of a time series provided in an embodiment of the present invention;

[0031] Figure 6B This is an optional schematic diagram of a time series provided in an embodiment of the present invention;

[0032] Figure 7 This is an optional schematic diagram of alarm based on time series provided in an embodiment of the present invention;

[0033] Figure 8 This is an optional schematic diagram of configuring alarm rules provided in an embodiment of the present invention;

[0034] Figure 9A This is an optional schematic diagram of batch alarm query provided in an embodiment of the present invention;

[0035] Figure 9B This is an optional schematic diagram of batch alarm query provided in an embodiment of the present invention;

[0036] Figure 10 This is an optional flowchart illustrating the business monitoring method provided in this embodiment of the invention;

[0037] Figure 11 This is an optional schematic diagram of the upper boundary, middle line, and lower boundary provided in an embodiment of the present invention;

[0038] Figure 12 This is an optional schematic diagram of a strip-shaped region provided in an embodiment of the present invention;

[0039] Figure 13 This is an optional schematic diagram of the fitting curve provided in an embodiment of the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on the present invention. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0042] In the following description, the terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first" and "second" may be interchanged in a specific order or sequence where permitted, so that the embodiments of the invention described herein can be implemented in an order other than that illustrated or described herein.

[0043] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein is for the purpose of describing embodiments of the invention only and is not intended to limit the invention.

[0044] In the implementation of this application, the collection and processing of relevant data should strictly comply with the requirements of relevant laws and regulations, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0045] Before providing a further detailed description of the embodiments of the present invention, the nouns and terms involved in the embodiments of the present invention will be explained, and the nouns and terms involved in the embodiments of the present invention shall be interpreted as follows.

[0046] 1) Data point sequence: Also known as a time series, it refers to a sequence of collected business data arranged in chronological order. Each data point in the data point sequence is similar to a key-value pair, including two dimensions of information: the collection time and the business data. For each fixed point in time, there is a unique business data value corresponding to it in the data point sequence.

[0047] 2) Volatility trend: also known as monotonicity, refers to the trend of a data point sequence over a certain period of time. The types of volatility trends can include normal trends and abnormal trends, or include normal trends, upward trends and downward trends.

[0048] 3) Business data: refers to the data obtained by the business party in carrying out business. The specific type depends on the actual application scenario. For example, it can be the success rate of interface, the success rate of accessing web pages, the buffering rate of live video, and the number of online users of the application.

[0049] 4) Artificial Intelligence (AI): This refers to the theories, methods, technologies, and application systems that utilize digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results. This invention does not limit the type of artificial intelligence model; for example, it could be a machine learning model.

[0050] 5) Big data refers to data sets that cannot be captured, managed, and processed using conventional software tools within a certain time frame. It is a massive, rapidly growing, and diverse information asset that requires new processing models to achieve stronger decision-making, insight discovery, and process optimization capabilities.

[0051] 6) Database: refers to a collection of data stored together in a certain way, which can be shared by multiple users, has the lowest possible redundancy, and is independent of the application. Users can perform operations such as adding, querying, updating, and deleting data in the database.

[0052] 7) Blockchain: An encrypted, chain-like storage structure for transactions formed by blocks.

[0053] 8) Blockchain Network: A collection of nodes that incorporate new blocks into a blockchain through consensus.

[0054] This invention provides a business monitoring method, apparatus, electronic device, and computer-readable storage medium, which can refine the monitoring granularity and improve the accuracy of business monitoring.

[0055] The following describes exemplary applications of the electronic device provided in the embodiments of the present invention. The electronic device provided in the embodiments of the present invention can be a server, such as a server deployed in the cloud, which performs business monitoring on the first data point sequence obtained remotely from the business party and feeds back the obtained business monitoring results to the business party, or sends alarm prompts to the business party based on the business monitoring results; it can also be a terminal device, such as the business party's own local monitoring device, which performs business monitoring based on the obtained first data point sequence; or it can even be a handheld terminal or other devices.

[0056] See Figure 1 , Figure 1This is an optional architecture diagram of a business monitoring system 100 that incorporates cloud technology provided in an embodiment of the present invention. Cloud technology refers to a hosting technology that unifies a series of resources such as hardware, software, and networks within a wide area network or local area network to realize the computation, storage, processing, and sharing of data. It can also be understood as a general term for network technology, information technology, integration technology, management platform technology, and application technology based on cloud computing business models.

[0057] like Figure 1 As shown, to support a business monitoring application, server 200 (exemplarily showing a server cluster consisting of servers 200-1, 200-2, and 200-3) runs the business according to specific business logic, that is, builds the production environment of the business. The business logic can be configured by terminal device 400 (exemplarily showing terminal device 400-1 and terminal device 400-2) through network 500-3, or it can be configured by other terminal devices. The business logic is such as the running logic of a web service or the running logic of an application.

[0058] In some embodiments, server 300 (exemplarily shown as a server cluster consisting of servers 300-1, 300-2, and 300-3) can execute the business monitoring method provided in this embodiment of the invention. Specifically, terminal device 400 connects to server 300 in the cloud via network 500-2, opening the data acquisition interface for the corresponding business in the production environment to server 300. Simultaneously, terminal device 400 sends alarm rules to server 300 according to the user's configuration in the cloud monitoring assistant. The alarm rules include, for example, issuing an alarm when the fluctuation trend of a certain data point matches a set alarm fluctuation trend. Thus, server 300 can collect business data from server 200 in real time through the data acquisition interface, forming a data point sequence, and determine whether to send an alarm notification based on the alarm rules. For example, in a big data scenario, servers 200-1 to 200-3 can form a distributed file system for storing business data in the production environment, and server 300 collects business data from the distributed file system through the data acquisition interface. Of course, in addition to collecting business data from server 200 in real time through the data acquisition interface, server 300 can also access databases or blockchains to obtain the business data stored on server 200, or the data point sequence built based on the business data. Here, server 200 can also store the business data or data point sequence in other scalable storage systems, not limited to databases and blockchains.

[0059] In addition, terminal device 400 can also execute the business monitoring method provided in this embodiment of the invention. Specifically, terminal device 400 collects business data from server 200 in real time through network 500-3, constructs a data point sequence, and determines whether to issue an alarm based on the alarm rules set by the user in the cloud monitoring assistant. Similarly, terminal device 400 can also obtain business data or pre-constructed data point sequences by accessing a database or blockchain. Figure 1 The network 500 shown (including networks 500-1, 500-2 and 500-3) can be a wide area network or a local area network, or a combination of both. Figure 1 The users shown can be business personnel, monitoring personnel who interface with business personnel, or other types of personnel.

[0060] Terminal device 400 can display various results during the business monitoring process in graphical interface 410 (graphical interfaces 410-1 and 410-2 are shown as examples). Figure 1 Taking the received alarm message as an example, the alarm message contains information about an anomaly at a certain data point.

[0061] It is worth noting that, in this embodiment of the invention, the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. The terminal device can be various types of terminal devices such as laptops, tablets, desktop computers, set-top boxes, and mobile devices (e.g., mobile phones, portable music players, personal digital assistants, dedicated messaging devices, portable gaming devices), but is not limited to these. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, and this embodiment of the invention does not impose any limitations on this connection.

[0062] This invention can also be implemented using blockchain technology. Blockchain is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include a blockchain underlying platform, a platform product service layer, and an application service layer.

[0063] The underlying blockchain platform can include processing modules such as user management, basic services, smart contracts, and operational monitoring. The user management module is responsible for managing the identity information of all blockchain participants, including maintaining public and private key generation (account management), key management, and maintaining the correspondence between user real identities and blockchain addresses (access management). Furthermore, under authorization, it monitors and audits transactions of certain real identities and provides risk control rule configuration (risk control audit). The basic services module is deployed on all blockchain node devices to verify the validity of business requests. After consensus is reached on valid requests, they are recorded in storage. For a new business request, the basic services first perform interface adaptation parsing and authentication (interface adaptation), and then encrypt the business information through a consensus algorithm (consensus management). After encryption, the data is transmitted completely and consistently to the shared ledger (network communication) and recorded and stored. The smart contract module is responsible for contract registration, issuance, triggering, and execution. Developers can define contract logic using a programming language and publish it to the blockchain (contract registration). According to the contract terms, the key or other events are invoked to trigger execution and complete the contract logic. It also provides functions for contract upgrades and cancellations. The operation monitoring module is mainly responsible for deployment, configuration modification, contract settings, cloud adaptation, and real-time status visualization output during product release, such as alarms, monitoring network conditions, and monitoring the health status of node devices.

[0064] See Figure 2 , Figure 2 This is an optional architecture diagram of a business monitoring system 110 combined with a blockchain network provided in an embodiment of the present invention, including a blockchain network 600 (nodes 610-1 to 610-3 are shown as an example), an authentication center 700, and an electronic device 800, which may be a server (e.g., Figure 1 The server 300 shown can also be a terminal device (e.g., Figure 1 The terminal device 400 shown is determined according to the actual application scenario, and will be explained in detail below.

[0065] The types of blockchain networks 600 are flexible and diverse, including public, private, and consortium blockchains. For example, with a public blockchain, any electronic device, such as a terminal device or server, can access the blockchain network 600 without authorization. With a consortium blockchain, electronic devices can access the blockchain network 600 after obtaining authorization, becoming a special type of node within the blockchain network—a client node.

[0066] It should be noted that client nodes may only provide functionality to support business systems in initiating transactions (e.g., for storing data on-chain or querying on-chain data). For the functions of native nodes of the blockchain network 600, such as the sorting function, consensus service, and ledger function mentioned below, client nodes may implement them by default or selectively (e.g., depending on the specific business needs of the business system). This allows for the maximum migration of data and business processing logic from electronic devices to the blockchain network 600, achieving trustworthiness and traceability of data and business processing through the blockchain network 600. The blockchain network 600 receives transactions submitted by client nodes and executes these transactions to update or query the ledger.

[0067] The following example illustrates an exemplary application of blockchain networks, using the example of an electronic device accessing a blockchain network to query data point sequences.

[0068] Electronic device 800 connects to blockchain network 600, becoming a client node of blockchain network 600. When electronic device 800 needs to query the data point sequence of a certain business party, it sends a query request, including the identification information of that business party, to the blockchain network in the form of a transaction. The transaction specifies the smart contract to be invoked to implement the query operation and the parameters to be passed to the smart contract. The transaction also carries a digital signature signed by electronic device 800 (for example, obtained by encrypting the transaction digest using electronic device 800's digital certificate) and broadcasts the transaction to blockchain network 600. The digital certificate can be obtained by electronic device 800 through registration with certification authority 700.

[0069] When node 610 in blockchain network 600 receives a transaction, it verifies the digital signature carried in the transaction. If the digital signature verification is successful, it checks whether the electronic device 800, carried in the transaction, has the authority to conduct the transaction based on its identity. Either verification of the digital signature or the authority verification will cause the transaction to fail. After successful verification, node 610 signs its own digital signature and continues to broadcast it within blockchain network 600.

[0070] After receiving a successfully verified transaction, node 610, which has sorting capabilities in blockchain network 600, populates the transaction into a new block and broadcasts it to the nodes in blockchain network 600 that provide consensus services.

[0071] In the blockchain network 600, nodes 610 providing consensus services conduct a consensus process for new blocks to reach agreement. Nodes providing ledger functionality append the new block to the end of the blockchain and execute the transactions in the new block. For transactions querying data point sequences, based on the business party's identification information carried in the transaction, the system queries the state database for the data point sequence corresponding to that identification information and sends the data point sequence to the electronic device 800. It is worth noting that the state database stores the data point sequences corresponding to the identification information in key-value pairs, and the data stored in the state database is usually the same as the data stored in the blockchain. When responding to query transactions, the system prioritizes responding based on the data in the state database, thereby improving response efficiency.

[0072] The process of adding the data point sequence to the chain is similar to the process described above, for example, Figure 1 The server 300 shown connects to the blockchain network 600, becoming a client node of the blockchain network 600. Then, the server 300 generates a transaction that submits a sequence of data points and broadcasts the transaction to the blockchain network 600. The transaction carries the identification information of the business party. After the node 610 in the blockchain network 600 verifies the transaction, fills the block, and reaches consensus, the node providing the ledger function appends the new block to the end of the blockchain and executes the transactions in the new block: for transactions that submit a sequence of data points, the identification information of the business party and the sequence of data points in the transaction are stored in the state database in the form of key-value pairs.

[0073] It's worth noting that blockchain can be used to store data point sequences specifically for a single business party, or it can store data point sequences for multiple business parties simultaneously. For the latter, multiple channels can be constructed within the blockchain, with each channel storing the data point sequence for one business party. When responding to a query transaction, node 610 first determines the corresponding channel based on the business party's identification information carried in the transaction, and then retrieves the data point sequence from that channel.

[0074] The following describes exemplary applications of the electronic device provided in the embodiments of the present invention. The electronic device can be implemented as a terminal device or as a server.

[0075] The following explanation uses an electronic device as an example of a server. (See also...) Figure 3 , Figure 3 The server 300 provided in this embodiment of the invention (for example, may be...) Figure 1 The diagram shown is an architecture diagram of server 300. Figure 3The server 300 shown includes at least one processor 310, memory 340, and at least one network interface 320. The various components of server 300 are coupled together via a bus system 330. It is understood that the bus system 330 is used to implement communication between these components. In addition to a data bus, the bus system 330 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 3 The general labeled all buses as Bus System 330.

[0076] The processor 310 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0077] The memory 340 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state storage, hard disk drives, optical disk drives, etc. The memory 340 may optionally include one or more storage devices physically located away from the processor 310.

[0078] The memory 340 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), and the volatile memory may be random access memory (RAM). The memory 340 described in this embodiment is intended to include any suitable type of memory.

[0079] In some embodiments, memory 340 is capable of storing data to support various operations, examples of which include programs, modules, and data structures or subsets or supersets thereof, as illustrated below.

[0080] Operating system 341 includes system programs for handling various basic system services and performing hardware-related tasks, such as the framework layer, core library layer, driver layer, etc., for implementing various basic business functions and handling hardware-based tasks;

[0081] The network communication module 342 is used to reach other computing devices via one or more (wired or wireless) network interfaces 320, such as Bluetooth, WiFi, and Universal Serial Bus (USB).

[0082] In some embodiments, the service monitoring device provided in this invention can be implemented in software. Figure 3 A business monitoring device 343 stored in memory 340 is shown. This device can be software in the form of programs and plug-ins, and includes the following software modules: an acquisition module 3431, an interception module 3432, a segmentation module 3433, and a comparison module 3434. These modules are logically connected and can therefore be arbitrarily combined or further split according to their implemented functions. The functions of each module will be described below.

[0083] In other embodiments, the service monitoring device provided in this invention can be implemented in hardware. As an example, the service monitoring device provided in this invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the service monitoring method provided in this invention. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0084] The service monitoring method provided in this embodiment of the invention can be executed by the aforementioned server, or by a terminal device (e.g., it could be...). Figure 1 The terminal devices 400-1 and 400-2 shown can be used to execute the commands, or the server and the terminal devices can be used together to execute the commands.

[0085] The following section will describe, in conjunction with the exemplary applications and structures of the electronic devices described above, the process of implementing a business monitoring method in an electronic device through an embedded business monitoring device.

[0086] See Figure 4 and Figure 5A , Figure 4 This is a schematic diagram of the architecture of the business monitoring device 343 provided in an embodiment of the present invention, illustrating the process of implementing business monitoring through a series of modules. Figure 5A This is a flowchart illustrating the business monitoring method provided in this embodiment of the invention, which will be combined with... Figure 4 right Figure 5A The steps shown are explained.

[0087] In step 101, the first data point sequence of the business party is obtained, wherein the first data point sequence includes multiple data points obtained by collecting data on the business party's business at multiple time points.

[0088] As an example, see Figure 4 In the acquisition module 3431, during the business operations of the business party, a sequence of data points is acquired. For ease of differentiation, this sequence is named the first data point sequence. This first data point sequence can be acquired in real-time or periodically. The first data point sequence includes multiple data points obtained from collecting data on the business party's operations at multiple time points; each data point includes information in two dimensions: collection time and business data. This embodiment of the invention does not limit the type of business data, such as the success rate of an interface, the success rate of accessing a webpage, the buffering rate of live video, and the number of online users of an application. Furthermore, it is worth noting that the first data point sequence can be updated in real-time. For example, the first data point sequence can include multiple data points obtained from real-time collection of data on the business operations since the business went live, where the latest data point in the first data point sequence is the data point collected from the business operations at the current time.

[0089] In some embodiments, the above-mentioned acquisition of the first data point sequence of the business party can be achieved by performing one of the following processes: obtaining the first data point sequence of the business party from the business party's database; sending a query request including the identification information of the business party to the blockchain network to obtain the first data point sequence stored in the blockchain that corresponds to the identification information of the business party.

[0090] As an example, see Figure 4 In the acquisition module 3431, the first data point sequence of the business party can be obtained in two ways. The first method is for the case where the first data point sequence of the business party is pre-stored in the business party's database. When performing business monitoring, the first data point sequence of the business party is obtained from the business party's database.

[0091] The second approach addresses the scenario where the business entity's first data point sequence is pre-stored on the blockchain. During business monitoring, a query request containing the business entity's identification information is sent to the blockchain network. Specifically, the transaction containing this query request is sent to a node on the blockchain network. After verification, block filling, and consensus, the node sends the first data point sequence stored on the blockchain, corresponding to the business entity's identification information, to the query request initiator. If a state database corresponding to the blockchain exists, the first data point sequence stored in the state database, corresponding to the business entity's identification information, is sent to the query request initiator first, improving response efficiency. This approach enhances the flexibility of obtaining the first data point sequence and is suitable for various storage scenarios.

[0092] In some embodiments, between any steps, the method further includes: obtaining a key from a business party; wherein the key from the business party is used by the business party to encrypt a first data point sequence, so as to store the encrypted first data point sequence in a blockchain; wherein the blockchain is used to store the first data point sequences of multiple business parties.

[0093] After sending a query request to the blockchain network, which includes the business party's identification information, the process also includes: decrypting the encrypted first data point sequence obtained from the blockchain based on the business party's key.

[0094] Here, the blockchain can be used specifically to store the first data point sequence of a single business party, or it can be used to store the first data point sequences of multiple business parties. For the latter, due to the public and transparent nature of the blockchain, to prevent data leakage, each business party can encrypt its first data point sequence using its own key before uploading it to the blockchain. In this way, the blockchain stores the encrypted first data point sequence. Even if unauthorized personnel obtain the encrypted first data point sequence from the blockchain, they cannot access the actual data because they do not possess the corresponding key.

[0095] During business monitoring, the system obtains the business party's key, retrieves the encrypted first data point sequence from the blockchain, and decrypts the encrypted first data point sequence using the key. It's worth noting that this embodiment of the invention does not limit the key generation method; for example, it can be generated using a symmetric encryption algorithm or an asymmetric encryption algorithm. This method enhances the security of the first data point sequence in the blockchain.

[0096] In step 102, starting from the data point to be monitored among multiple data points, multiple data points that were collected before the starting point and within the first set time period are extracted from the first data point sequence. The data point to be monitored and the extracted multiple data points are then used to form the second data point sequence.

[0097] As an example, see Figure 4 In the interception module 3432, according to pre-set monitoring rules, the data points to be monitored in the first data point sequence are determined. Starting from these data points, multiple data points whose acquisition time is before the starting point and falls within a first set time period are intercepted from the first data point sequence. The data points to be monitored and the intercepted data points constitute a second data point sequence. Here, the first set time period can be set according to the actual application scenario, such as being set to several minutes.

[0098] In some embodiments, before step 102, the method further includes performing one of the following processes to determine the data points to be monitored: obtaining a set time to be monitored, and determining data points in the first data point sequence whose acquisition time matches the time to be monitored, as data points to be monitored; wherein the time to be monitored is a time point or a time period; determining data points in the first data point sequence whose acquisition time matches the current time point, as data points to be monitored.

[0099] This invention provides two methods for determining data points to be monitored. The first method involves obtaining a manually set monitoring time and identifying data points in a first data point sequence whose collection time matches the monitoring time. These data points are then used as monitoring data points. The monitoring time can be a specific time point or a time period. The monitoring time can be set according to the business monitoring needs of the business entity. For example, if the business entity's peak usage time is from 3 PM to 4 PM, which is also the time when the business entity generates significant revenue, then the monitoring time can be set to 3 PM to 4 PM daily for targeted monitoring. This way, other less important time periods can be left unmonitored, saving computing resources.

[0100] The second approach involves identifying data points in the first data point sequence whose collection time matches the current time point. These data points are then used as the monitoring data points, enabling real-time monitoring and immediate feedback when business anomalies occur. This approach enhances the flexibility of business monitoring, allowing users to choose either method based on their specific application scenario.

[0101] In some embodiments, a monitoring mode can be set, and all data points collected during the monitoring mode can be used as data points to be monitored, that is, real-time monitoring is performed when the monitoring mode is on; while data points collected during the monitoring mode is off are not processed.

[0102] The control options for turning the monitoring mode on and off can be presented in Figure 1 The graphical interface 410 of the terminal device 400, as shown, is presented, for example, in a cloud monitoring assistant. Thus, relevant personnel on the business side can enable or disable monitoring mode through control options based on the actual business situation. For example, if the business side has multiple monitoring personnel with varying shift schedules, each monitoring personnel can manually enable monitoring mode when it's their turn and manually disable it when their shift ends. As another example, if a significant problem arises in the business at any given time, relevant personnel can enable monitoring mode for real-time monitoring to determine if business data has been affected. By setting monitoring modes, the flexibility of monitoring is improved from another perspective.

[0103] In step 103, the second data point sequence is subjected to boundary division processing to obtain the upper limit and lower limit of the business data of the data points to be monitored.

[0104] Here, based on the changes in multiple data points in the second data point sequence, the second data point sequence is delineated by boundaries. For example, boundary delineation is performed using statistical algorithms to obtain the upper and lower limits of the business data for the data points to be monitored.

[0105] In step 104, the business data of the data point to be monitored is compared with the upper limit and lower limit of the business data to obtain the fluctuation trend of the data point to be monitored, which serves as the business monitoring result for the business party.

[0106] As an example, see Figure 4 In comparison module 3434, the business data of the data point to be monitored is compared with the upper and lower limits of the business data. Based on the comparison results, the fluctuation trend of the data point to be monitored is determined, and this fluctuation trend is the business monitoring result for the business party. On this basis, it can also be combined with the set alarm rules to determine whether to issue an alarm based on the fluctuation trend.

[0107] In some embodiments, the above-mentioned comparison of the business data of the data point to be monitored with the upper and lower limits of the business data can be implemented in the following way to obtain the fluctuation trend of the data point to be monitored: when the value of the business data of the data point to be monitored is greater than the upper limit of the business data, the fluctuation trend of the data point to be monitored is determined to be an upward trend; when the value of the business data of the data point to be monitored is less than the lower limit of the business data, the fluctuation trend of the data point to be monitored is determined to be a downward trend; when the business data of the data point to be monitored is less than or equal to the upper limit of the business data and greater than or equal to the lower limit of the business data, the fluctuation trend of the data point to be monitored is determined to be a normal trend.

[0108] Here, the upper and lower limits of business data define the normal range of values ​​for the business data of the monitored data points. Specifically, when the business data of a monitored data point is less than or equal to the upper limit and greater than or equal to the lower limit, its fluctuation trend is considered normal. Otherwise, it is considered abnormal. Further, this can be subdivided: when the value of the business data is greater than the upper limit, the fluctuation trend is considered upward; when the value is less than the lower limit, the fluctuation trend is considered downward.

[0109] In some embodiments, after step 104, the method further includes: when the fluctuation trend of the data point to be monitored matches the alarm fluctuation trend, sending an alarm notification to the business party based on the data point to be monitored.

[0110] Here, the alarm rule can be a specifically defined alarm fluctuation trend. This trend can be an abnormal trend, or a more specific upward or downward trend, depending on the actual application scenario. For example, when the business data is the number of online users of the application, the alarm fluctuation trend is set to a downward trend; when the business data is the failure rate of an interface, the alarm fluctuation trend is set to an upward trend. When the fluctuation trend of the data point to be monitored, determined in step 104, matches the alarm fluctuation trend, an alarm notification is sent to the business entity based on the data point to be monitored. The method of sending the alarm notification is not limited here; it can include, but is not limited to, SMS, telephone, and email.

[0111] It's worth noting that, in addition to alarm fluctuation trends, other settings can be configured in the alarm rules. For example, constraints can be set for the year-on-year increase / decrease ratio, month-on-month increase / decrease ratio, and the business data itself of the monitored data points. An alarm is sent when these constraints are not met. This approach helps remind business personnel to further analyze the causes of the anomalies.

[0112] In some embodiments, between any steps, the method further includes: performing one of the following processes to obtain an alarm fluctuation trend: acquiring a set alarm fluctuation trend; determining the data points in the first data point sequence that have sent alarm notifications, and determining the fluctuation trend of the data points that have sent alarm notifications as the alarm fluctuation trend.

[0113] In addition to manually setting alarm fluctuation trends, this embodiment of the invention can also identify data points in the first data point sequence that have sent alarm notifications. These data points can be determined manually or based on content in the alarm rules other than alarm fluctuation trends. Then, the data points that have sent alarm notifications are identified as data points to be monitored. After processing in steps 102 to 104, the fluctuation trend of these data points is obtained. If the fluctuation trend of the data point is a normal trend, no processing is performed; if the fluctuation trend of the data point is a trend other than the normal trend, then the fluctuation trend of the data point is identified as an alarm fluctuation trend. This method improves the flexibility of determining alarm fluctuation trends, allowing for automatic determination of alarm fluctuation trends by combining historical alarm records.

[0114] Through the embodiments of the invention, for Figure 5A As can be seen from the above exemplary implementation, the embodiments of the present invention use data points in the data point sequence as the finest monitoring granularity, which improves the accuracy of business monitoring and enables timely feedback when business is abnormal.

[0115] In some embodiments, see Figure 5B , Figure 5B This is an optional flowchart illustrating the business monitoring method provided in an embodiment of the present invention. Figure 5A Step 103 shown can be implemented through steps 201 to 203, or through steps 204 to 205, which will be explained in conjunction with each step.

[0116] In step 201, the mean and standard deviation of the business data of multiple data points in the second data point sequence are determined.

[0117] As an example, see Figure 4 In the partitioning module 3433, for the intercepted second data point sequence, the mean and standard deviation of the business data of multiple data points in the second data point sequence are determined. Since the standard deviation reflects the dispersion of data points in the second data point sequence, the upper limit and lower limit of business data can be determined based on the mean and standard deviation.

[0118] In some embodiments, before step 201, the method further includes: traversing the data points in the second data point sequence and performing the following processing on the traversed data points: extracting data points in the second data point sequence whose acquisition time is before the traversed data points and which are within a second set time period; constructing a third data point sequence from the traversed data points and the extracted data points in the second data point sequence; and averaging the service data of all data points in the third data point sequence to obtain the average service data of the traversed data points.

[0119] The above-mentioned determination of the mean and standard deviation of business data of multiple data points in the second data point sequence can be achieved in the following way: determine the mean of business data of multiple data points in the second data point sequence, and determine the standard deviation of the average business data of multiple data points in the second data point sequence.

[0120] In addition to directly determining the mean and standard deviation of the business data of all data points in the third data point sequence, this embodiment of the invention also provides another method. Specifically, all data points in the second data point sequence are traversed, and data points whose collection time is before the traversed data points and fall within a second predetermined time period are extracted from the second data point sequence. The traversed data points and the extracted data points from the second data point sequence constitute the third data point sequence, wherein the second predetermined time period is shorter than the first predetermined time period and can be manually set. For the first data point in the second data point sequence, since there are no other data points before this data point, this data point itself is taken as part of the third data point sequence.

[0121] After obtaining the third data point sequence, the business data of all data points in the third data point sequence is averaged to obtain the average business data of the traversed data points. After the traversal is complete, the average business data of each data point in the third data point sequence can be obtained. Then, the mean of the business data of multiple data points in the second data point sequence is determined, and the standard deviation of the average business data of multiple data points in the second data point sequence is also determined. In this way, by combining the changes in data points within a short period (a second set time period), the accuracy of the upper and lower limits of the subsequently obtained business data can be improved.

[0122] In some embodiments, before step 201, the method further includes: traversing the data points in the second data point sequence and performing the following processing on the traversed data points: when the traversed data point is the first data point in the second data point sequence, keeping the service data of the traversed data point unchanged as the average service data of the traversed data point; when the traversed data point is not the first data point in the second data point sequence, weighting the service data of the traversed data point and the average service data of the previous data point according to the respective weight parameters of the traversed data point and the previous data point to obtain the average service data of the traversed data point; wherein, the sum of the weight parameters of the traversed data point and the weight parameters of the previous data point is 1;

[0123] The above-mentioned determination of the mean and standard deviation of business data of multiple data points in the second data point sequence can be achieved in the following way: determine the mean of business data of multiple data points in the second data point sequence, and determine the standard deviation of the average business data of multiple data points in the second data point sequence.

[0124] Besides the method of intercepting the third data point sequence, this embodiment of the invention can also improve the accuracy of the subsequently obtained upper and lower limits of business data through another method. Specifically, all data points in the second data point sequence are traversed. When the traversed data point is the first data point in the second data point sequence, the business data of the traversed data point remains unchanged and is used as the average business data of the traversed data point. When the traversed data point is not the first data point in the second data point sequence, the business data of the traversed data point and the average business data of the previous data point are weighted and summed according to the weight parameters of the traversed data point and the previous data point to obtain the average business data of the traversed data point. The sum of the weight parameters of the traversed data point and the previous data point is 1.

[0125] Similarly, after obtaining the average business data of each data point in the second data point sequence, the mean of the business data of multiple data points in the second data point sequence is determined, and the standard deviation of the average business data of multiple data points in the second data point sequence is determined.

[0126] In step 202, the standard deviation is multiplied by the set parameter.

[0127] Here, the parameters can be set according to the actual application scenario; for example, the parameter can be set to 3.

[0128] In step 203, the result of the product processing is added to the mean to obtain the upper limit of the business data of the data point to be monitored, and the mean is subtracted from the result of the product processing to obtain the lower limit of the business data of the data point to be monitored.

[0129] The product result obtained in step 202 reflects the normal fluctuation range of the data point to be monitored based on the mean. Therefore, the product result is added to the mean to obtain the upper limit of the business data of the data point to be monitored, and the mean is subtracted from the product result to obtain the lower limit of the business data of the data point to be monitored. In this way, the fluctuation trend of the data point to be monitored can be determined based on the upper and lower limits of the business data.

[0130] exist Figure 5B middle, Figure 5A Step 103 shown can also be implemented through steps 204 to 205. In step 204, in the two-dimensional spatial distribution of the business data values ​​with time on the horizontal axis and business data values ​​on the vertical axis, a fitting curve that can fit the second data point sequence is determined, and the fitting business data corresponding to the data points to be monitored in the fitting curve is determined.

[0131] As an example, see Figure 4In the partitioning module 3433, a polynomial fitting is performed on the second data point sequence. Specifically, in a two-dimensional spatial distribution where the horizontal axis represents time and the vertical axis represents the numerical values ​​of the business data, a fitting curve that can fit the second data point sequence is determined. Then, based on the acquisition time of the data points to be monitored, the corresponding fitted business data is found in the obtained fitting curve.

[0132] In step 205, the values ​​of the fitted business data are increased according to a set ratio to obtain the upper limit of the business data of the data points to be monitored, and the values ​​of the fitted business data are decreased according to a set ratio to obtain the lower limit of the business data of the data points to be monitored.

[0133] After obtaining the fitted business data corresponding to the data point to be monitored, the value of the fitted business data is increased according to a set ratio to obtain the upper limit of the business data for the data point to be monitored; the value of the fitted business data is then decreased according to the same set ratio to obtain the lower limit of the business data for the data point to be monitored. The set ratio is, for example, 10% or other values.

[0134] Through the embodiments of the invention, for Figure 5B As can be seen from the above exemplary implementation, in the embodiments of the present invention, the upper limit and lower limit of business data can be obtained by calculating the mean and standard deviation, or by using polynomial fitting, thereby improving flexibility.

[0135] In some embodiments, see Figure 5C , Figure 5C This is an optional flowchart illustrating the business monitoring method provided in this embodiment of the invention, based on... Figure 5A Before step 101, in step 301, a sequence of sample data points including sample data points and corresponding sample fluctuation trends can be obtained; wherein, the type of sample fluctuation trend includes normal trend and abnormal trend, or includes normal trend, upward trend and downward trend.

[0136] Here, the sample data point sequence can be a data point sequence of a business entity acquired at a certain point in history. The sample data point sequence and the first data point sequence correspond to the same business entity, and the number of data points included in the sample data point sequence is the same as that in the first data point sequence. The sample data point sequence includes at least one sample data point labeled with a sample fluctuation trend. This sample fluctuation trend is labeled manually or through other means, and the type of sample fluctuation trend includes normal trends and abnormal trends, or includes normal trends, upward trends, and downward trends.

[0137] In step 302, the sample data point sequence is classified using an artificial intelligence model to obtain the fluctuation trend of the sample data points to be compared.

[0138] In this embodiment of the invention, the fluctuation trend of the data points to be monitored can be obtained through an artificial intelligence model. Prior to this, the artificial intelligence model needs to be trained. Specifically, the artificial intelligence model performs forward propagation processing on the sample data point sequence, i.e., classification processing, to obtain the fluctuation trend to be compared for each data point in the sample data point sequence. Taking the artificial intelligence model as an example, which includes an input layer, a hidden layer, and an output layer, the forward propagation processing refers to inputting the sample data point sequence into the input layer, and then processing it sequentially through the input layer, hidden layer, and output layer, ultimately obtaining the fluctuation trend to be compared for each data point in the sample data point sequence output by the output layer.

[0139] When the sample fluctuation trend includes both normal and abnormal trends, the AI ​​model is a binary classification model; when the sample fluctuation trend includes normal, upward, and downward trends, the AI ​​model is a three-class classification model. Furthermore, to improve the training performance of the AI ​​model, the number of sample data points corresponding to different sample fluctuation trends can be kept as balanced as possible. Balance here does not mean complete equality, but rather that they are within a set proportional range. For example, the ratio between the number of sample data points corresponding to normal trends and the number of sample data points corresponding to abnormal trends should be greater than 1 / 4 and less than 2.

[0140] In step 303, backpropagation is performed in the artificial intelligence model based on the difference between the fluctuation trend of the sample data points to be compared and the fluctuation trend of the sample, and the weight parameters of the artificial intelligence model are updated during the backpropagation process.

[0141] After obtaining the fluctuation trend of the sample data points to be compared in step 302, the difference between the fluctuation trend of the sample data points to be compared and the fluctuation trend of the sample data points is determined according to the loss function of the artificial intelligence model. This difference is the loss value. The type of loss function is not limited; for example, it can be a cross-entropy loss function. Based on the obtained difference, backpropagation is performed in the artificial intelligence model, and during backpropagation, the weight parameters of the artificial intelligence model are updated along the gradient descent direction. Again, taking a neural network model including an input layer, hidden layers, and an output layer as an example, backpropagation means propagating the obtained difference sequentially to the output layer, hidden layer, and input layer. During the propagation to each layer, the gradient is calculated based on the difference, and the weight parameters of that layer are updated along the gradient descent direction.

[0142] based on Figure 5A After step 102, in step 304, the second data point sequence can be classified and processed by an artificial intelligence model to obtain the fluctuation trend of the data points to be monitored, which can be used as the business monitoring result for the business party.

[0143] Here, the second data point sequence is classified using an artificial intelligence model to obtain the fluctuation trend of each data point in the second data point sequence, that is, to obtain the fluctuation trend of the data point to be monitored. In this embodiment of the invention, the artificial intelligence model can be an unsupervised model or a supervised model. Unsupervised models include isolated forest models and one-class support vector machines (OneClass SVM models), while supervised models include neural network models.

[0144] If the artificial intelligence model is unsupervised, there is no need to train the artificial intelligence model; that is, step 304 can be executed directly. If the artificial intelligence model is supervised, steps 301 to 303 are executed first, and then in step 304, the trained artificial intelligence model is used to classify the second data point sequence.

[0145] Through the embodiments of the invention, for Figure 5C As can be seen from the above exemplary implementation, in addition to the boundary division processing method, the embodiments of the present invention also provide a method of obtaining fluctuation trends through artificial intelligence models, thereby improving the flexibility of business monitoring.

[0146] The following describes an exemplary application of the present invention in a practical application scenario. After the business goes live, business data in the production environment is collected through a data acquisition system and arranged in chronological order to obtain a time series. At each fixed moment, each time series corresponds to a unique value, and the data points in the time series are similar to key-value pairs. Typically, in a time series, the time interval between two adjacent data points is a constant value (e.g., 1 second, 10 seconds, 1 minute, 5 minutes, or 1 hour). For ease of understanding, the present invention provides, as shown in the following... Figure 6A and Figure 6B The diagram shown illustrates the time series. Figure 6A and Figure 6B In this process, one monitoring data point reported every minute by the data acquisition system is obtained, and the data points are arranged to obtain a time series. Figure 6A and Figure 6B The horizontal axis represents the data collection time, and the vertical axis represents business data. The types of business data include, but are not limited to, the success rate of interfaces, the success rate of accessing web pages, the buffering rate of live video, and the number of online users of the application.

[0147] In business monitoring, alerts are typically involved. An alert refers to the issuance of a warning based on pre-configured alert rules when a significant anomaly (sudden increase or decrease) occurs in the time series data reported by the business entity. Alert rules can vary widely, such as a 10% year-on-year increase, a 10% month-on-month decrease, or a data point value less than 99%. Once a data point in the time series meets the alert rules, an alert is sent to the business entity's personnel. This invention provides, for example... Figure 7 The diagram shown illustrates alarms based on time series data. Figure 7 If the alarm rule is set to the case that the value of the business data of a data point is less than 98%, then an alarm will be triggered during the time period from 17:06 to 18:03 if the value of the business data of a data point is less than 98%.

[0148] In the field of monitoring, business users are more concerned with the trend of individual data points over time. Therefore, in this embodiment of the invention, monotonicity analysis is performed on the time series, that is, analyzing whether the time series shows an upward, downward, or normal fluctuation trend over a certain period of time, to achieve business monitoring with data points as the finest granularity. In other words, the business monitoring in this embodiment of the invention includes two aspects: First, configuring detection for the business data itself, such as configuring an alarm in the alarm rules when the value of a certain business data is greater than or less than a certain threshold; second, performing monotonicity judgment based on data points, and determining whether to issue an alarm based on the monotonicity conclusion. For the first aspect, this embodiment of the invention provides, as follows: Figure 8 The diagram shown illustrates the configuration of alarm rules (i.e., triggering conditions). Alarm rules include, but are not limited to, logical AND, logical OR, and logical NOT checks. Figure 8 The code shows that an alarm is triggered when the business data is greater than or equal to 1. Here, the duration of the alarm rule can also be set, such as for a statistical period, which is the time interval mentioned above. Figure 8 The example used is 1 minute. For the second aspect, users still need to configure the corresponding alarm rules, but unlike the first aspect which includes the configuration of various thresholds, here it is only necessary to configure "upward alarm", "downward alarm" or "alarm for both upward and downward" for the time series, and the configuration process is relatively simple.

[0149] In embodiments of the present invention, alarm methods include, but are not limited to, instant messaging notifications, SMS, email, and telephone. In addition, embodiments of the present invention can also provide an alarm query function on a web page, that is, based on a user's query request on the web page including a time point or time period, displaying business data for that time point or time period, and indicating whether an alarm has been issued. Embodiments of the present invention provide, for example... Figure 9A and Figure 9B The diagram shown illustrates the batch alarm query process. Figure 9A and Figure 9B When a user queries data points that triggered alarms today and yesterday, the web interface displays the business data of those data points that triggered alarms today (represented by dashed lines) and the business data of those data points that triggered alarms yesterday (represented by solid lines). Of course, in addition to yesterday, more dates can be displayed based on the query request, such as the business data of data points that triggered alarms a week ago. Figure 9A The business data shown in the time series are Figure 9B The types of business data shown in the time series are different. Furthermore, in Figure 9A and Figure 9B The document also shows options for labeling samples as positive and negative, facilitating model training based on user-labeled positive and negative samples.

[0150] The embodiments of the present invention provide, as follows Figure 10 The flowchart of the business monitoring method shown is combined with... Figure 10 The specific process of business monitoring may include the following steps:

[0151] 1) Data storage: This mainly involves a database, which is used to store multiple time series. Each time series stores a type of business data, and the time series corresponds to the first data point sequence mentioned above.

[0152] 2) Data Extraction: For each time series in the database, extract the data points within the past n minutes, starting from the current moment. The data point at the current moment corresponds to the data point to be monitored mentioned above, and n minutes corresponds to the first set time period mentioned above.

[0153] 3) Algorithm Analysis: Multiple algorithms are used to comprehensively analyze the time series extracted in step 2) (corresponding to the second data point sequence mentioned above). The overall algorithm structure is sequential, as described below:

[0154] Step 1: First, use the control chart algorithm to obtain the fluctuation trend of the data points at the current moment. If there is an anomaly (successfully matching the alarm rule), return the result directly; if there is no anomaly, use the multinomial fitting algorithm, i.e., proceed to Step 2.

[0155] Step 2: If the multinomial fitting algorithm produces anomalies, return the results directly; if it produces normal results, analyze them using an artificial intelligence model, proceeding to Step 3.

[0156] Step 3: Use an artificial intelligence model to obtain the results and return them directly.

[0157] 4) Match the results obtained in step 3) with the alarm rules previously configured by the user based on business experience, such as "alarm for rising", "alarm for falling", or "alarm for both rising and falling".

[0158] 5) If the matching result in step 4) is a successful match, an alarm will be sent to the user; if the matching result in step 4) is a failed match, no alarm will be sent.

[0159] The following section details each algorithm. For ease of explanation, we will use {x1,…,x} as the model name. n} represents a time series of length n, where the last business data x n This represents the business data collected at the current moment, x1 to x2. n-1 This represents historically collected business data, where n is an integer greater than 1.

[0160] 1) Control chart algorithm.

[0161] Control chart algorithms include multiple sub-algorithms. You can choose to use one of them, or you can combine multiple sub-algorithms in a serial manner. They are described below in numbered form:

[0162] ① Determine the time series {x1,…,x} n The mean μ and variance σ of multiple business data in} 2 :

[0163]

[0164] Furthermore, we obtain the upper bound (UCL, Upper Center Line), the middle line (Center Line), and the lower bound (LCL, Lower Center Line):

[0165] UCL=μ+L·σ

[0166] CenteLine = μ

[0167] LCL=μ-L·σ

[0168] Where L corresponds to the setting parameter mentioned above, which can be 3. The upper bound corresponds to the upper limit of the business data mentioned above, and the lower bound corresponds to the lower limit of the business data mentioned above. When x n When >UCL, it means x n The corresponding data points show an upward trend; when x n <UCL, indicating x n The corresponding data points show a downward trend; if x n ∈[LCL,UCL], indicating that x n The corresponding data points are in a stable state.

[0169] ② Iterate through i from 1 to n. In each iteration, use a window of length w to cover the time series {x1,…,xn}. n Business data x in} i and x i The previous business data is used to obtain a new time series (corresponding to the third data point sequence mentioned above), where i is an integer. It is worth noting that during the traversal, there may be cases where the length of the new time series is less than w. Specifically, when i < w, the length of the new time series is i, that is, it includes i business data points; when i ≥ w, the length of the new time series is w.

[0170] For the obtained new time series, calculate the mean of all business data within it to obtain x. i Average business data M w (i), that is

[0171] The variance can be calculated based on the obtained average business data. Therefore, the upper bound, intermediate line, and lower bound obtained based on algorithm ② are:

[0172]

[0173] Cente Line = μ

[0174]

[0175] ③ For the time series {x1,…,x} n Update each piece of business data in}:

[0176]

[0177] Where λ is the weight coefficient (corresponding to the weight parameters of the data points traversed above), and λ∈(0,1).

[0178] Further analysis can yield the variance. That is to say Therefore, the upper bound, intermediate line, and lower bound obtained based on algorithm ③ are:

[0179]

[0180] Cente Line = μ

[0181]

[0182] For ease of understanding, Figure 11The diagram illustrates the upper bound, median line, and lower bound obtained using a control chart algorithm. Furthermore, if for each data point in the original time series of the database, historical data points spanning n minutes are extracted, and a control chart algorithm is executed to obtain the upper bound, median line, and lower bound for each data point in the original time series, a band-shaped interval composed of upper and lower bounds that continuously change with time points can ultimately be obtained. This invention provides, as shown in the embodiments below. Figure 12 The schematic diagram of the strip-shaped region shown is in Figure 12 In this case, n is 20 and L is 2.

[0183] 2) Polynomial fitting.

[0184] Polynomial fitting uses a polynomial equation to fit a time series {x1,…,x}. n Since a continuous function on a closed interval can be uniformly approximated by a polynomial series, polynomial fitting can be performed on data points composed of various types of business data to obtain a fitted curve. The polynomial equation is of the form: P(x) = a m x m +a m-1 x m-1 +…+a1x+a0, this equation represents a polynomial of degree m. Embodiments of this invention provide, as follows: Figure 13 The schematic diagram of the fitted curve shown is in Figure 13 In the time series {x1,…,x} n The figures are represented by solid lines, and the fitted curves are represented by dashed lines. For the obtained fitted curves, a percentage can be set to increase the vertical axis to obtain the upper bound; a percentage can be set to decrease the vertical axis to obtain the lower bound. The percentage can be set to, for example, 10%, 20%, or other percentages.

[0185] 3) Artificial intelligence model.

[0186] Besides algorithms 1) and 2), artificial intelligence models can also be used to determine the fluctuation trends of the data points to be monitored. Artificial intelligence models, such as unsupervised algorithms in machine learning (e.g., Isolation Forest and One Class SVM models), can also be used to complete the training and deployment of supervised models through labeling. Labeling, for example... Figure 9A and Figure 9B The labeled positive and negative samples shown (corresponding to a supervised binary classification model) can also be labels of three types, including normal trend, upward trend and downward trend (corresponding to a supervised ternary classification model).

[0187] It is worth noting that algorithms 1), 2), and 3) above can be used as follows: Figure 10 As shown, it can be used in a serial structure, combined arbitrarily, or used independently during business monitoring.

[0188] The following technical effects can be achieved through the above methods:

[0189] 1) It reduces the cost of manual maintenance. In the field of monitoring, the effect is that it is no longer necessary to repeatedly configure various parameters and thresholds for each time series. The overall monitoring system can be maintained with less manpower.

[0190] 2) The solutions provided by related technologies can only determine whether the overall time series shows an upward trend, a normal trend, or a downward trend. However, the embodiments of the present invention improve the accuracy of business monitoring by detecting the monotonicity of the time series with data points as the finest granularity.

[0191] 3) It can detect time series that are normally distributed, as well as time series that are not normally distributed, making it highly applicable.

[0192] 4) It can be put into production in real time and can also be connected to new business data (time series) in real time, with strong robustness and real-time performance.

[0193] The following continues to describe an exemplary structure of the business monitoring device 343 provided in the embodiments of the present invention as a software module. In some embodiments, such as Figure 3 As shown, the software modules stored in the business monitoring device 343 in the memory 340 may include: an acquisition module 3431, used to acquire a first data point sequence of the business party, wherein the first data point sequence includes multiple data points corresponding to the collection of the business party's business at multiple time points; an extraction module 3432, used to extract multiple data points from the first data point sequence that are collected before the starting point and within a first set time period, taking the data point to be monitored among the multiple data points as the starting point, and forming a second data point sequence with the data point to be monitored and the extracted multiple data points; a division module 3433, used to perform boundary division processing on the second data point sequence to obtain the upper limit and lower limit of the business data of the data point to be monitored; and a comparison module 3434, used to compare the business data of the data point to be monitored with the upper limit and lower limit of the business data to obtain the fluctuation trend of the data point to be monitored, as the business monitoring result of the business party.

[0194] In some embodiments, the partitioning module 3433 is further configured to: determine the mean and standard deviation of the business data of multiple data points in the second data point sequence; multiply the standard deviation with a set parameter; add the result of the multiplication to the mean to obtain the upper limit of the business data of the data point to be monitored, and subtract the result of the multiplication from the mean to obtain the lower limit of the business data of the data point to be monitored.

[0195] In some embodiments, the business monitoring device 343 further includes: a first traversal module, configured to traverse the data points in the second data point sequence and perform the following processing on the traversed data points: extracting data points in the second data point sequence whose collection time is before the traversed data points and which are within a second set time period; constructing a third data point sequence from the traversed data points and the extracted data points in the second data point sequence; and averaging the business data of all data points in the third data point sequence to obtain the average business data of the traversed data points.

[0196] The partitioning module 3433 is also used to: determine the mean of the business data of multiple data points in the second data point sequence, and determine the standard deviation of the average business data of multiple data points in the second data point sequence.

[0197] In some embodiments, the service monitoring device 343 further includes: a second traversal module, configured to traverse the data points in the second data point sequence and perform the following processing on the traversed data points: when the traversed data point is the first data point in the second data point sequence, the service data of the traversed data point is kept unchanged as the average service data of the traversed data point; when the traversed data point is not the first data point in the second data point sequence, the service data of the traversed data point and the average service data of the previous data point are weighted according to the weight parameters of the traversed data point and the previous data point to obtain the average service data of the traversed data point; wherein the sum of the weight parameters of the traversed data point and the weight parameters of the previous data point is 1;

[0198] The partitioning module 3433 is also used to: determine the mean of the business data of multiple data points in the second data point sequence, and determine the standard deviation of the average business data of multiple data points in the second data point sequence.

[0199] In some embodiments, the partitioning module 3433 is further configured to: determine a fitting curve that can fit the second data point sequence in a two-dimensional spatial distribution where the horizontal axis is time and the vertical axis is the numerical value of the business data, and determine the fitting business data in the fitting curve that corresponds to the data point to be monitored; adjust the value of the fitting business data upward according to a set ratio to obtain the upper limit of the business data of the data point to be monitored, and adjust the value of the fitting business data downward according to a set ratio to obtain the lower limit of the business data of the data point to be monitored.

[0200] In some embodiments, the comparison module 3434 is further configured to: determine that the fluctuation trend of the monitored data point is an upward trend when the value of the business data of the monitored data point is greater than the upper limit of the business data; determine that the fluctuation trend of the monitored data point is a downward trend when the value of the business data of the monitored data point is less than the lower limit of the business data; and determine that the fluctuation trend of the monitored data point is a normal trend when the business data of the monitored data point is less than or equal to the upper limit of the business data and greater than or equal to the lower limit of the business data.

[0201] In some embodiments, the business monitoring device 343 further includes a classification module, used to classify the second data point sequence using an artificial intelligence model to obtain the fluctuation trend of the data points to be monitored.

[0202] In some embodiments, the business monitoring device 343 further includes an alarm module, which is used to send an alarm notification to the business party based on the data point to be monitored when the fluctuation trend of the data point to be monitored matches the alarm fluctuation trend.

[0203] In some embodiments, the business monitoring device 343 further includes: a sample acquisition module, used to acquire a sequence of sample data points including sample data points and corresponding sample fluctuation trends; a sample classification module, used to classify the sequence of sample data points using an artificial intelligence model to obtain the fluctuation trends of the sample data points to be compared; and a training module, used to perform backpropagation in the artificial intelligence model based on the difference between the fluctuation trends of the sample data points to be compared and the sample fluctuation trends, and to update the weight parameters of the artificial intelligence model during the backpropagation process; wherein the type of sample fluctuation trend includes normal trends and abnormal trends, or includes normal trends, upward trends and downward trends.

[0204] In some embodiments, the business monitoring device 343 further includes: a rule acquisition module, configured to perform one of the following processes to obtain an alarm fluctuation trend: acquire a set alarm fluctuation trend; determine the data points in the first data point sequence that have sent alarm prompts, and determine the fluctuation trend of the data points that have sent alarm prompts as the alarm fluctuation trend.

[0205] In some embodiments, the business monitoring device 343 further includes: a data point determination module, configured to perform one of the following processes to determine the data points to be monitored: acquiring a set time to be monitored, and determining data points in the first data point sequence whose acquisition time matches the time to be monitored, as data points to be monitored; wherein the time to be monitored is a time point or a time period; determining data points in the first data point sequence whose acquisition time matches the current time point, as data points to be monitored.

[0206] In some embodiments, the acquisition module 3431 is configured to perform one of the following processes: acquire a first data point sequence of the business party from the business party's database; send a query request including the business party's identification information to the blockchain network to acquire the first data point sequence stored in the blockchain that corresponds to the business party's identification information.

[0207] In some embodiments, the business monitoring device 343 further includes: a key acquisition module, used to acquire the key of the business party; wherein the key of the business party is used by the business party to encrypt the first data point sequence, so as to store the encrypted first data point sequence in the blockchain; wherein the blockchain is used to store the first data point sequences of multiple business parties;

[0208] The business monitoring device 343 also includes a decryption module, used to decrypt the encrypted first data point sequence obtained from the blockchain based on the business party's key.

[0209] This invention provides a computer-readable storage medium storing executable instructions. When these executable instructions are executed by a processor, they cause the processor to perform the method provided in this invention, for example... Figure 5A , Figure 5B or Figure 5C The business monitoring method is shown. It is worth noting that computers include various computing devices, including terminal devices and servers.

[0210] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EP ROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a variety of devices including one or any combination of the above-mentioned memories.

[0211] In some embodiments, executable instructions may take the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0212] As an example, executable instructions may, but do not necessarily, correspond to files in a file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a Hyper Text Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple co-located files (e.g., files that store one or more modules, subroutines, or code sections).

[0213] As an example, executable instructions can be deployed to execute on a single computing device, or on multiple computing devices located in one location, or on multiple computing devices distributed across multiple locations and interconnected via a communication network.

[0214] In summary, the following technical effects can be achieved through the embodiments of the present invention:

[0215] 1) The solutions provided by related technologies can only determine whether the overall data point sequence shows an upward trend, a normal trend, or a downward trend. However, the embodiments of the present invention improve the accuracy of business monitoring by detecting the monotonicity of the data point sequence with data points as the finest granularity, and can provide timely feedback when business is abnormal.

[0216] 2) Reduced manual maintenance costs: alarm fluctuation trends only need to be configured in the alarm rules, and the monitoring system can be maintained with relatively low manpower costs.

[0217] 3) During business monitoring, a single algorithm can be used, or multiple algorithms can be used in sequence; at the same time, business monitoring can be implemented in the cloud or locally, which is highly flexible.

[0218] 4) It can detect data point sequences that are normally distributed, as well as data point sequences that are not normally distributed, making it highly applicable.

[0219] 5) It can be put into production in real time and can also access new business data (data point sequence) in real time, with strong robustness and real-time performance; in addition, it can be applied to a variety of application scenarios such as cloud scenarios and big data scenarios.

[0220] The above are merely embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and scope of the present invention are included within the scope of protection of the present invention.

Claims

1. A business monitoring method, characterized in that, include: Obtain the first data point sequence of the business party, wherein the first data point sequence includes multiple data points obtained by collecting data on the business party's business at multiple time points; Starting from the data point to be monitored among the multiple data points, multiple data points that were collected before the starting point and are within a first set time period are extracted from the first data point sequence. The data point to be monitored and the extracted multiple data points constitute a second data point sequence. Traverse the data points in the second data point sequence and perform the following processing on the traversed data points: extract data points in the second data point sequence whose collection time is before the traversed data points and which are within the second set time period, and combine the traversed data points and the extracted data points in the second data point sequence to form a third data point sequence; The average business data of all data points in the third data point sequence is averaged to obtain the average business data of the traversed data points. Determine the mean of the business data of multiple data points in the second data point sequence, and determine the standard deviation of the average business data of multiple data points in the second data point sequence; The standard deviation is multiplied by the set parameter; the result of the product is added to the mean to obtain the upper limit of the business data of the data point to be monitored; and the mean is subtracted from the result of the product to obtain the lower limit of the business data of the data point to be monitored. The business data of the data point to be monitored is compared with the upper limit and lower limit of the business data to obtain the fluctuation trend of the data point to be monitored, which serves as the business monitoring result for the business party.

2. The business monitoring method according to claim 1, characterized in that, The business monitoring method also includes: Iterate through the data points in the second data point sequence and perform the following processing on the traversed data points: When the traversed data point is the first data point in the second data point sequence, the business data of the traversed data point remains unchanged, and is used as the average business data of the traversed data point. When the traversed data point is not the first data point in the second data point sequence, the business data of the traversed data point and the average business data of the previous data point are weighted according to the weight parameters of the traversed data point and the previous data point to obtain the average business data of the traversed data point. Wherein, the sum of the weight parameter of the traversed data point and the weight parameter of the previous data point is 1.

3. The business monitoring method according to claim 1, characterized in that, The business monitoring method also includes: In a two-dimensional spatial distribution where the horizontal axis represents time and the vertical axis represents the numerical values ​​of the business data, a fitting curve that can fit the second data point sequence is determined, and the fitted business data corresponding to the data point to be monitored in the fitting curve is determined. The values ​​of the fitted business data are increased according to a set ratio to obtain the upper limit of the business data for the data points to be monitored. The values ​​of the fitted business data are adjusted downwards according to the set ratio to obtain the lower limit of the business data for the data points to be monitored.

4. The business monitoring method according to claim 1, characterized in that, The step of comparing the business data of the data point to be monitored with the upper limit and the lower limit of the business data to obtain the fluctuation trend of the data point to be monitored includes: When the value of the business data of the data point to be monitored is greater than the upper limit of the business data, the fluctuation trend of the data point to be monitored is determined to be an upward trend. When the value of the business data of the data point to be monitored is less than the lower limit of the business data, the fluctuation trend of the data point to be monitored is determined to be a downward trend. When the business data of the monitored data point is less than or equal to the upper limit of the business data and greater than or equal to the lower limit of the business data, the fluctuation trend of the monitored data point is determined to be a normal trend.

5. The business monitoring method according to any one of claims 1 to 4, characterized in that, After constructing the second data point sequence from the data points to be monitored and the multiple data points extracted, the method further includes: The fluctuation trend of the data points to be monitored is obtained by classifying the second data point sequence using an artificial intelligence model. The business monitoring method also includes: When the fluctuation trend of the monitored data point matches the alarm fluctuation trend, an alarm notification is sent to the business party based on the monitored data point.

6. The business monitoring method according to claim 5, characterized in that, Also includes: Obtain a sequence of sample data points, including sample data points and corresponding sample fluctuation trends; The artificial intelligence model is used to classify the sequence of sample data points to obtain the fluctuation trend of the sample data points to be compared. Based on the difference between the fluctuation trend of the sample data points to be compared and the sample fluctuation trend, backpropagation is performed in the artificial intelligence model, and During backpropagation, the weight parameters of the artificial intelligence model are updated; The types of fluctuation trends in the samples include normal trends and abnormal trends, or include normal trends, upward trends and downward trends.

7. The business monitoring method according to claim 5, characterized in that, Also includes: Perform one of the following processes to obtain the alarm fluctuation trend: Obtain the set alarm fluctuation trend; Identify the data points in the first data point sequence that have received alarm notifications, and The fluctuation trend of the data points for which alarm notifications have been sent is determined as the alarm fluctuation trend.

8. The business monitoring method according to any one of claims 1 to 4, characterized in that, Also includes: Perform one of the following processes to determine the data point to be monitored: Get the set monitoring time, and The data points in the first data point sequence whose collection time matches the time to be monitored are identified as the data points to be monitored; wherein, the time to be monitored is a time point or a time period; The data points in the first data point sequence whose collection time matches the current time point are identified and used as the data points to be monitored.

9. The business monitoring method according to any one of claims 1 to 4, characterized in that, The acquisition of the first data point sequence from the business party includes: Perform one of the following processes: Obtain the first data point sequence of the business party from the business party's database; Send a query request to the blockchain network, including the identification information of the business party, to obtain the first data point sequence stored in the blockchain that corresponds to the identification information of the business party.

10. The business monitoring method according to claim 9, characterized in that, The business monitoring method also includes: Obtain the key of the business party; wherein the key of the business party is used by the business party to encrypt the first data point sequence, so as to store the encrypted first data point sequence in the blockchain; The blockchain is used to store the first data point sequence of multiple business parties; After sending a query request including the business party's identification information to the blockchain network, the method further includes: Based on the key of the business party, the encrypted sequence of the first data points obtained from the blockchain is decrypted.

11. A business monitoring device, characterized in that, include: The acquisition module is used to acquire a first data point sequence of the business party, wherein the first data point sequence includes multiple data points obtained by collecting data from the business party's business at multiple time points. The interception module is used to take the data point to be monitored among the multiple data points as the starting point, and intercept multiple data points in the first data point sequence whose collection time is before the starting point and are within a first set time period, and to form a second data point sequence by the data point to be monitored and the intercepted multiple data points. The segmentation module is used to traverse the data points in the second data point sequence and perform the following processing on the traversed data points: Extract data points from the second data point sequence whose acquisition time is before the traversed data points and falls within a second set time period; combine the traversed data points and the extracted data points from the second data point sequence to form a third data point sequence; average the business data of all data points in the third data point sequence to obtain the average business data of the traversed data points; determine the mean of the business data of multiple data points in the second data point sequence and determine the standard deviation of the average business data of multiple data points in the second data point sequence; multiply the standard deviation by a set parameter; add the result of the product to the mean to obtain the upper limit of the business data of the data point to be monitored, and subtract the result of the product from the mean to obtain the lower limit of the business data of the data point to be monitored. The comparison module is used to compare the business data of the data point to be monitored with the upper limit and the lower limit of the business data to obtain the fluctuation trend of the data point to be monitored, which serves as the business monitoring result for the business party.

12. The business monitoring device according to claim 11, characterized in that, The business monitoring device also includes: The second traversal module is used to traverse the data points in the second data point sequence and perform the following processing on the traversed data points: when the traversed data point is the first data point in the second data point sequence, the business data of the traversed data point remains unchanged and is used as the average business data of the traversed data point; when the traversed data point is not the first data point in the second data point sequence, the business data of the traversed data point and the average business data of the previous data point are weighted according to the weight parameters of the traversed data point and the previous data point to obtain the average business data of the traversed data point; wherein, the sum of the weight parameters of the traversed data point and the weight parameters of the previous data point is 1.

13. The business monitoring device according to claim 11, characterized in that, The segmentation module is further configured to determine, in a two-dimensional spatial distribution of the business data values ​​with time as the horizontal axis and the business data values ​​as the vertical axis, a fitting curve that can fit the second data point sequence, and to determine the fitted business data corresponding to the data point to be monitored in the fitted curve; to increase the value of the fitted business data according to a set ratio to obtain the upper limit of the business data of the data point to be monitored, and to decrease the value of the fitted business data according to the set ratio to obtain the lower limit of the business data of the data point to be monitored.

14. The business monitoring device according to claim 11, characterized in that, The comparison module is also used to determine that the fluctuation trend of the monitored data point is an upward trend when the value of the business data of the monitored data point is greater than the upper limit of the business data. When the value of the business data of the data point to be monitored is less than the lower limit of the business data, the fluctuation trend of the data point to be monitored is determined to be a downward trend. When the business data of the monitored data point is less than or equal to the upper limit of the business data and greater than or equal to the lower limit of the business data, the fluctuation trend of the monitored data point is determined to be a normal trend.

15. The business monitoring device according to any one of claims 11 to 14, characterized in that, The business monitoring device also includes: The classification module is used to classify the second data point sequence using an artificial intelligence model to obtain the fluctuation trend of the data points to be monitored. The alarm module is used to send an alarm notification to the business party when the fluctuation trend of the monitored data point matches the alarm fluctuation trend.

16. The business monitoring device according to claim 15, characterized in that, The business monitoring device also includes: The sample acquisition module is used to acquire a sequence of sample data points, including sample data points and corresponding sample fluctuation trends; classify the sequence of sample data points using the artificial intelligence model to obtain the fluctuation trends of the sample data points to be compared; perform backpropagation in the artificial intelligence model based on the difference between the fluctuation trends of the sample data points to be compared and the sample fluctuation trends, and update the weight parameters of the artificial intelligence model during the backpropagation process; wherein, the type of sample fluctuation trend includes normal trends and abnormal trends, or includes normal trends, upward trends and downward trends.

17. The business monitoring device according to claim 15, characterized in that, The business monitoring device also includes: The rule acquisition module is used to perform one of the following processes to obtain the alarm fluctuation trend: acquire the set alarm fluctuation trend; determine the data points in the first data point sequence that have sent alarm prompts, and determine the fluctuation trend of the data points that have sent alarm prompts as the alarm fluctuation trend.

18. The business monitoring device according to any one of claims 11 to 14, characterized in that, The business monitoring device also includes: The data point determination module is used to perform one of the following processes to determine the data point to be monitored: obtaining a set monitoring time and determining data points in the first data point sequence whose acquisition time matches the monitoring time, as the data point to be monitored; wherein the monitoring time is a time point or a time period; determining data points in the first data point sequence whose acquisition time matches the current time point, as the data point to be monitored.

19. The business monitoring device according to any one of claims 11 to 14, characterized in that, The acquisition module is further configured to perform one of the following processes: acquire the first data point sequence of the business party from the business party's database; send a query request including the identification information of the business party to the blockchain network to acquire the first data point sequence stored in the blockchain that corresponds to the identification information of the business party.

20. The business monitoring device according to claim 19, characterized in that, The business monitoring device also includes: A key acquisition module is used to acquire the key of the business party; wherein the key of the business party is used by the business party to encrypt the first data point sequence, so as to store the encrypted first data point sequence in the blockchain; wherein the blockchain is used to store the first data point sequences of multiple business parties; The decryption module is used to decrypt the encrypted first data point sequence obtained from the blockchain based on the key of the business party.

21. An electronic device, characterized in that, include: Memory, used to store executable instructions; A processor, when executing executable instructions stored in the memory, implements the service monitoring method according to any one of claims 1 to 10.

22. A computer-readable storage medium, characterized in that, It stores executable instructions for causing the processor to execute, thereby implementing the business monitoring method according to any one of claims 1 to 10.

Images