CN110611694A

CN110611694A - Data processing center based on virtualization master-slave container

Info

Publication number: CN110611694A
Application number: CN201910377639.4A
Authority: CN
Inventors: 秦锡忠
Original assignee: Hangzhou Hengyu Culture And Art Planning Co Ltd
Current assignee: Hangzhou Hengyu Culture And Art Planning Co Ltd
Priority date: 2019-04-29
Filing date: 2019-04-29
Publication date: 2019-12-24

Abstract

A data processing center based on a virtualization master-slave container forms a master control center and an auxiliary control center of a cloud computing system platform of our company and is an industry and enterprise oriented informatization system. In the prior art, the problems of cloud computing information security, network delay or network interruption are mostly solved by storage data processing, and the data processing mode wastes excessive data processing time to cause network delay or interruption. Aiming at the defects of the prior art, the data processing center based on the virtualized master-slave container is created and invented, the data processing center is mainly characterized by matrixing data processing, a data processing technology characterized by instruction execution under a virtual device is formed, the virtualized master-slave container is used for high-quality processing of data imported by an application server, and the data processing center under a virtualized environment is formed. Can be applied to the fields: large 3D digital authoring; a domestic backbone communication network; a system requiring confidentiality in China; the fields of big agriculture + + and the like.

Description

Data processing center based on virtualization master-slave container

Technical Field

The invention provides a cloud data integration processing algorithm, and particularly relates to a high-quality data processing method and device based on a master container and a slave container in a virtualization environment.

In the system architecture, a master container and a slave container are master control centers of a platform; the application server is an auxiliary control center of the platform, the main control center is responsible for managing and supervising each auxiliary control center, and the auxiliary control center becomes a regional information processing and transaction service center under the supervision and management of the main control center. The main control center and the auxiliary control center form a cloud computing system platform.

Background

With the advent of the cloud era, big data has attracted more and more attention. The big data is used as the vocabulary of the IT industry which is the most hot at present, and the utilization of the commercial value of the big data, such as data warehouse, data security, data analysis, data mining and the like, becomes the profit focus which is gradually pursued by the industry people. The large data is the large amount of unstructured and semi-structured data generated when data operation is applied. Large data analysis is often linked to cloud computing, and real-time large data set analysis requires allocation of work to tens, hundreds, or even thousands of computers to process the vast amount of data involved, achieving capture, management, processing, and consolidation in a reasonable time into sufficient and advantageous resources. At present, people talk about big data technology and big data application most, and the big data technology, big data engineering, big data science, big data application and other fields can be divided. Large data requires special techniques to efficiently process large amounts of data that are tolerant of elapsed time. In the prior art, the technology suitable for large data includes massive parallel processing databases, data mining, distributed file systems, distributed databases, cloud computing platforms, the internet and scalable storage systems.

The strategic significance of big data technology is not to grasp huge data information, but to specialize the data with significance. In other words, if data is compared to an industry, the key to realizing profitability in such an industry is to improve the "processing ability" of the data and to realize "value-added" of the data by "processing".

With the gradual rise of the internet of things, stronger data analysis and machine learning are performed on large data applications aiming at some challenges, and the new architecture also comes up endlessly. The application of a brand-new cluster can foresee that, with the continuous development of technologies such as cloud computing and edge computing, the realization of future insights by utilizing the analysis capability of big data will gradually become practical.

The industrialization and the informatization are mutually fused, so that elements such as knowledge, science and technology, culture, art and the like are more integrated into products and services, the products and the services more embody sociality, vitality and environmental harmony, and the human brain and the computer are beautiful innovation points of companies; the network marketing system has the characteristics of the network marketing system under the network economic condition, and shows more characteristics of universality, instantaneity, economy, interactivity and the like compared with the traditional marketing; the informatization system integrates marketing theory, soft marketing theory and direct and repeated marketing theory, and the theories deduct and create a new theoretical framework of network marketing from the aspects of combination of marketing strategy and internet technology, psychological change trend of consumers, direct and repeated marketing and the like. The network marketing is greatly developed, and meanwhile, the traditional marketing is transformed into a large-scale and industrialized service industry from aspects of marketing strategy, promotion strategy, marketing organization and the like, the method takes creation, creation and innovation as fundamental means, takes cultural content and creative achievement as core values, takes intellectual property realization or consumption as transaction characteristics, and provides an industry cluster with internal connection for the social public with cultural experience. The existing computing informatization technology is far from meeting the requirements of cloud algorithms, and the modern cloud computing needs a high-quality data processing technology, so that the informatization construction is improved to a new milestone.

Regarding cloud computing, the cloud computing is a distributed technology, a virtualization technology and a service technology, and has a plurality of characteristics and advantages. (1) The method comprises the steps of (1) super-large scale, (2) abstraction, (3) high reliability, (4) universality, (5) high expandability, (6) on-demand service, (7) low price, (8) automation, (9) energy conservation and environmental protection, and (10) a perfect operation and maintenance mechanism. Because of the features and advantages, cloud computing can provide users with more convenient experience and lower cost, and the features and advantages are one of the reasons why cloud computing can stand out and can be advocated by most people in the industry.

With respect to cloud deployment, the patterns are: public clouds, private clouds, and hybrid clouds. The three modes have very different requirements on platform management. For users, the control of resource sharing, the requirement for system efficiency, and the cost investment budget of enterprises are not the same, and the cloud computing system scale and manageability required by the enterprises are also different. Therefore, the cloud computing platform management scheme needs to consider customization requirements more and can meet application requirements of different scenes.

The 'horizontal appearance' of cloud computing allows many people to regard the cloud computing as a brand-new technology, but in fact, the cloud computing has many defects, (1) a huge system needs huge capital investment, (2) operation and maintenance are difficult, (3) information security problems, (4) high energy consumption, and (5) network delay or interruption.

In the prior art, the problems of cloud computing information security, network delay or network interruption are mostly solved by taking stored data processing as a characteristic, and the data processing method is realized by a master server and a slave server, so that excessive data processing time is wasted, and the network delay or interruption is caused.

Disclosure of Invention

Aiming at the defects of the prior art, the data processing center based on the virtualization master-slave container is created and invented. The method is organically coupled with an application server by mainly taking matrixing data processing as a characteristic, and forms a characteristic data processing technology implemented by an instruction under a virtual device. The data processing technology is a virtual environment formed by a main control center and an auxiliary control center, and a virtualized main container and a virtualized slave container are used for high-quality processing of data imported by an application server, so that the data processing center in the virtual environment is formed. The technical scheme is satisfied by using virtualization technology and container technology, technical innovation is further performed, and a data processing center based on a virtualization master-slave container is generated under the technical background, and the core invention points of the data processing center are as follows: 1. the private cloud deployment enables users to operate on respective channels without mutual interference; 2. the application of the encryption technology is that a coder is added at a sending end, a decoder is correspondingly needed at a receiving end, the digital baseband signal is artificially disturbed (encrypted), and decryption is carried out at the receiving end; 3. the application of the tangent-plane enterprise bus technology in digital communication solves the problems of 'bit synchronization' or 'code element synchronization', 'group synchronization' or 'frame synchronization', solves the important problem of all 'synchronization' based on the data processing of a virtualized master container and a virtualized slave container, and more importantly, the data synchronization of the master container and the slave container are mutually backed up, so that a platform is safer and more reliable; 4. the network consists of an input layer, a plurality of hidden layers and an output layer, the SMP + AMP + BMP multi-core CPU processing mode realizes virtualization application, and the master control center with superior performance can monitor the operation of the whole system in real time; 5. the linearization communication technology, the multi-thread program design fully utilizes the processing capability of the CPU, and can be executed in parallel with high quality and high reliability, so that the reasonable performance on a platform system is improved; 6. the system can adapt to the environment, summarize the rule and complete certain operation, identification or process control; 7. in the data sequence program, basic elements are firstly converted into two-dimensional elements, and finally three-dimensional data elements are formed and self-adapted into space-time data elements; 8. the matrixing operation utilizes the coupling and synchronization principle, adopts the tight coupling and loose coupling and remote calling communication technology, and enables the master control center and the auxiliary control center to exert the optimal performance in internal communication application.

A data processing center software module based on a virtualization master-slave container mainly comprises: the system comprises a main container module, a slave container module, an input module, a session module, an output module, a container and application server interface module, an encryption module and an application server module.

1. Main container module

A main container module:

(1) and storing XML files defined by EJB-JAR, WAR, EAR and MBean.

(2) Remote interface (remote interface): service methods of session beans are defined, and the methods can be received by a local interface (local interface) from an EJB container, and the local interface allows direct memory interaction between the beans.

(3) Remote or local interface Bean class (Bean class): the bean class contains the business logic.

(4) The session Bean is used for realizing service logic and is divided into stateful Bean and stateless Bean.

(5) And the entity Bean represents the data of the real object.

(6) Message driven beans, MDBs, are components designed to specifically handle message based requests. It can send and receive asynchronous JMS messages and can easily interact with other EJBs.

2. From the container mould

The slave container has the same function and function as the master container, is contained by the master container, serves as the traditional container function in the prior art, can only run a single user process, is in charge of data synchronization and diary synchronization with the master container, and is in charge of integrating user data into the master container to carry out high-quality operation during deep operation.

3. Input module

The input module comprises a user input module and a big data interception module, wherein the big data interception is a data model and an interception mode which are made in advance by a result wanted by a user and is usually composed of some entity beans, a data model and a database.

4. Conversation module

The conversation module is composed of a large number of simple basic elements, and is a self-adaptive nonlinear dynamic system formed by the mutual connection of the elements. The structure and function of the element are simple, but the system behavior generated by the combination of the elements is very complex, and the aspects of the composition principle, the functional characteristics and the like are more close to the human brain. It does not perform operations step by step according to a given program, but can adapt itself to the environment, summarize the law, perform some kind of operation, recognition or process control.

The conversation element also has preliminary self-adapting and self-organizing ability. To meet the requirements of the surrounding environment. The same network may have different functions due to different learning modes and contents. The conversation element is a system with learning ability, the system can automatically find environmental characteristics and regularity, and has the function of being more similar to the human brain, and on the data sequence program, the basic element is firstly converted into a two-dimensional element, and finally, a three-dimensional data element is formed and is self-adapted into a four-dimensional data element of a space-time element. All these sessions are based on matrix calculations. And the whole conversation meta-process is realized on the data model by utilizing the super strong memory function of the computer.

5. Output module

The output module is the inverse process of the input module, is coupled with the input module to form an input port and an output port, and is also the result of big data interception.

6. Interface module for master-slave container and application server

The container and the application server interface adopts a tight coupling and loose coupling and remote calling mode to interface by utilizing a coupling and synchronization principle.

7. Encryption module

The module adopts a mode of SSL + digital signature to apply PKI technology to an information system. In this system, digital certificate based SSL secure information channel technology and digital signature technology. The application of the digital signature is taken as the main content. The authenticity of the user identity is ensured by using the digital certificate for authentication; the integrity of form data and the non-repudiation of the behavior of a main body are ensured by using the digital signature; the data encryption technology based on the digital certificate is utilized to ensure the safe transmission of the sensitive information, and a more reasonable, more convenient and more perfect safety platform is built.

8. Application server module

The application server module is completely software, the software application server is completely embedded into the main control center to form an auxiliary control center, and by utilizing the characteristics of low embedded technology cost, low power consumption, strong processing capability and the like, a unified operation interface and an application programming interface are provided for various file systems, and an interface for a user, a graphical interface, a library function API and the like are provided upwards; providing interfaces for interacting with hardware equipment, hardware drivers and the like downwards, and managing complex system resources; meanwhile, the method has more distinct characteristics in the aspects of system real-time performance, hardware dependence, software curing property, application specificity and the like. Functionally it can facilitate access to data sources and machine learning algorithms, such as clustering, regression, classification, graph patterns, and optimization, and provide a visual module. It contains a library of data matrices for storing and processing any type, and is able to handle very large matrices, even when they cannot be written to memory. Not only are many algorithms and tools provided, but interfaces are provided with other machine learning and data mining packages.

The application server submodule is as follows: a java virtual machine module; a JDWP virtual machine debugging module; a java data structure parser module; an application server and container interface module; and a middleware module.

1. Java virtual machine module

A virtual machine is an abstract computer, and is implemented by simulating various computer functions on an actual computer. The Java virtual machine has its own complete hardware architecture, such as processor, stack, registers, etc., and has a corresponding instruction system, which is capable of executing class files. From the operating system perspective, it is just a process, and a plurality of files of class complete a specific program function, and a plurality of bins consisting of class are a virtual machine.

2. JDWP virtual machine debugging module

The virtual machine is provided with a customized function to achieve the purpose of utilization, so that the virtual machine needs a debugging module, and the virtual machine debugging is that the test performance does not reach the standard, and is convenient to utilize. The parameters to be tested are:

3. java data structure parser module

The java data structure interpreter is a translator. Are the translation of constants, variables, data structures into instructions that a machine can execute. The java data structure interpreter performs machine layout on data items such as a data source, a machine learning algorithm, an access flag, a Class index, a parent Class index and the like, for example, a constant pool, a field table set, a method table set and an attribute table set, wherein fields and methods all have own attributes, and Class also has corresponding attributes, so that the field table set and the method table set are analyzed while the attribute table set is analyzed. The explanation execution process is carried out in three parts: loading of the code, checking of the code and execution of the code. The Java parser module design goal is to provide a computer model based on abstract specification descriptions, ensuring that Java code can run on any system that meets the specification.

4. Application server and container interface module

The application server and the container interface adopt a tight coupling and loose coupling remote calling mode to interface by utilizing a coupling and synchronization principle.

5. Middleware module

Middleware is a type of computer software that connects software components and applications, and includes a set of services to facilitate the interaction of multiple pieces of software running on one or more machines over a network. The interoperability provided by the technology promotes the evolution of a consistent distributed system architecture which is generally used for supporting and simplifying complex distributed application programs, middleware is arranged on an operating system, a network and a database, and the application software is arranged on the lower layer, and the overall function is to provide an environment for running and developing the application software on the upper layer, so as to help users flexibly and efficiently develop and integrate the complex application software.

The connection sequence of the application server sub-modules is as follows: firstly, establishing a java virtual machine; secondly, debugging the virtual machine; thirdly, explaining the virtual machine; fourthly, deploying middleware; and fifthly, the application server is interfaced with the container. The middleware is introduced to the third party technology, and is not described here.

Drawings

In order to simplify the technical description, the following further description is made with reference to the accompanying drawings in the specification, wherein arabic numerals 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 in fig. 1-15 constitute hardware devices or virtual devices, and the same arabic numerals denote the same devices or virtual devices; the Arabic numerals and the circles indicate steps, the same Arabic numerals and the circles indicate the same steps, the Arabic numerals and the circles are only used for distinguishing the steps, and the sequence program is not provided and is clearly indicated in the specification. Firstly, the step of-sixth is a visible step,the steps are implicit abstract machine implementation steps,it includes the following steps. The steps ofThe method is an operation method cluster for realizing one input layer, a plurality of hidden layers and one output layer by a machine.

Fig. 1 shows the entire platform system execution flow.

FIG. 2 illustrates platform system connection timing relationships.

Fig. 3 shows a connection relationship between the primary control center and the secondary control center of the platform system.

Fig. 4 shows the connection relationship between the master container and the slave container of the platform system master control center.

FIG. 5 illustrates a platform system single user flow.

FIG. 6 shows the matrix eigenvalues and eigenvectors of the structure of step two, data entry, step five, data exit and their intermediate hiding steps.

FIG. 7 shows the relationship between the matrix eigenvalue and eigenvector coordinate of the hiding step in the middle of the data entry step, the data exit step.

Fig. 8 shows the structure of the bus server 4 and step (iv).

Fig. 9 shows the steps performed by the platform system following the prior art standardization techniques, which is a supplementary explanation to fig. 8.

Fig. 10 shows a process in which the application server 2 loads a virtual machine and an application program.

Fig. 11 shows an application server 2 virtual machine debugging schematic.

FIG. 12 shows a java data structure interpreter implementation data structure schematic.

FIG. 13 illustrates a java data structure interpreter implementing a machine instruction schematic.

Fig. 14 shows a matrix eigenvalue to eigenvector coordinate relationship for the application server 2.

Fig. 15 shows a flow of steps implemented by the virtual machines of the primary control center and the secondary control center of the platform system.

Detailed Description

The following detailed description, taken in conjunction with the accompanying drawingsThe above module operation methods are introduced in the steps to illustrate a basic operation principle of a data processing center based on a virtualization master-slave container, where the data processing center based on the virtualization master-slave container is configured to include: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, the connection order thereof: the data processing system of which the outlet of 1 is the inlet of 2, the outlet of 2 is the inlet of 3, 4, the outlet of 3, 4 is the inlet of 8, the outlet of 8 is the inlet of 9, executes the steps of: the data processing method of r, c and c, storing computer program (instruction) on it, characterized by that when the program (instruction) is executed by processor it can implement r, and (5) carrying out the following steps.

Method and device for realizing figure 1

Fig. 1 is a diagram of a platform system. Wherein 1 is an application client; 2 is an application server; 3 is a server of the main control center; and 4, a server registration center of the main control center. In fig. 1, a client requests an application server, which is a user data user. The process that the application server requests the server 3 and requests the service registration center 4 is an entrance of user data user.home; step three, the main control center server 3 requests a process to the server registration center 4 to carry out tight coupling and loose coupling, remote calling, matrix operation and input and output data; step four, 4, the server registration center checks correctly, carries out WSDL and UDDI processing, and simultaneously responds to the main control center server 3 and the application servers 2 and 4 which are also called enterprise buses, and step four is a @ WebServices enterprise bus section interface, which is used for carrying out deep integration on @ WebService, carrying out Session Services integration on Session Services and realizing the processing of the three-dimensional data pool Session Services; step five, binding data through an output port and sending the binding data to the application server 2; step (c) is to send the calculation result to the client 1.

Second, FIG. 2 and method and apparatus for implementing the same

In fig. 2, 1 is a client section, a data entry, and a data exit; 2, an application server section, also called an auxiliary control center section, a data dividing inlet and a data outlet; 3, a master control center section, a data dividing inlet and a data outlet; 4, a server registration center, also called a virtual bus server, a data entry and a data exit; and 5, an SMP + AMP + BMP multi-core CPU processor virtual environment. In fig. 2, the exit of 1 is the entry of 2, the exit of 2 is the entry of 3, 4, and the exit of 4 is the session slice of step # SessionServices. The third step is the same as the fourth step in fig. 1, the seventh step is the @ WebService section interface, and the seventh step is the SessionServices session section.

Step (c) source code fragment:

the user data user dir is connected with the application server 3 through the source codes, wherein File is new File ("C: \ \ Users \ \ admin \ \ Workspace \ \ MyEclipse CI \ \ oshy client \ \ bin"); the method comprises the steps that a virtual machine is realized, and a new is realized to update the virtual machine; caching user data by a client through a StringBuilder sb ═ new StringBuilder () method; connecting to matrix operation through extensions CreatePathConfig, and coupling with the main control center; all parameters are delivered to the application server 3 by a main () method.

Step eight embodiment source code fragment:

the SessionService is a basic unit of a SessionServices session pool, and a source code implementation fragment:

a SessionService client in the SessionService code segment is new SessionService (); the coupling and synchronization of the main control center and the application server 2 are realized, and the main control center is tightly coupled and loosely coupled with the application server 2 in the step ninthly, the remote calling and the matrix operation are carried out.

Step ninthly, source code fragment of the embodiment:

in the source code, submitted transaction objects are identified by a unique tag ID, forming an application-level dictionary object pool Session services.

The process of using tight coupling and loose coupling, remote calling and matrix operation to realize the SessionServices of the dictionary object pool at the application program level comprises the following steps:

firstly, converting description into a two-dimensional matrix, firstly transmitting references to Keys, Values and Session services in a source code, then performing two-dimensional matrix conversion on the description element, integrating the description of the two-dimensional matrix data element into a bus server 4@ WebServices interface, and listing WebServices, wherein the listed WebServices are the bandwidth of an enterprise bus and the concurrency maximum value.

Secondly, after the description element matrix is converted, the following codes are used:

and (4) conveying to SessionServices for three-dimensional conversion, wherein the example realizes the three-dimensional conversion by inheriting an AbstractCoreObject parent class. Abstract CoreObject source code fragment:

and thirdly, connecting the matrix characteristic value and the matrix characteristic vector to a matrix operation library of a third party through instances CoreObject, and storing the operation result into a container 9.

Third, FIG. 3 implementation method and apparatus

In fig. 3, 1, 2, 3 are the same as fig. 3; 6 is a slave container; 7 is a main container; 10 is an interceptor; in fig. 3, the procedure (r), (c) is identical to fig. 2, and in the procedure (r) the large data is intercepted; step (ii) ofIs a database connection; A. b, C, D, E is a multi-user section with values equal to user1, user2, user3, user4, user.; and fourthly, after the task of the first-order calculation is finished, continuing the second-order calculation, the third-order calculation and the nth-order calculation.Is the application server 2 implementation step.

Fourth, FIG. 4 implementation method and apparatus

Fig. 4 is a connection diagram of the master and slave containers, in fig. 4, 2, 3, 4, 6, 7 are the same as in fig. 3, wherein the container 7 comprises the devices 4, 6, 8, 9; the container 8 is the container of the step eight and is used for realizing the three-dimensional data SessionServices processing pool; the container 9 is a container of step ninthly, and is used for realizing eigenvalue and eigenvector operation and eigenvalue decomposition of the data element description; in the step (c) of fig. 5, (c) is identical to that of fig. 4, (c) A, B, C, D, E is a multi-user section with values equal to user1, user2, user3, user4 and user. And step three, calculating a matrix.Is the application server 2 implementing the steps.

The main container implementation process comprises: step two, the operation result is put in the device 4 from the step four; the fourth step is to put the operation result in the container 8; step (III), (III) and (III) placing the operational result in a container 9; and fourthly, after the first-order operation task is completed, continuing second-order operation, third-order operation and nth-order operation.

Main container embodiment source code fragment:

the code segments are tightly coupled and loosely coupled, remotely called, matrix operated and remotely interfaced to the main control center server 3 by the application server 2 through the interface Lookup of the namespace technology.

Master center server 3 embodiment source code fragment:

this code is used to receive application server 2 inputs, "WebServices" assigned to Keys, "WebService" assigned to Values, "Session Services" assigned to Session Services, and then integrated into the description in preparation for the step of SessinServices operations.

Fifth, FIG. 5 implementation method and apparatus

Fig. 5 is a flow chart of a single user of a platform system, in fig. 5, 1 is an application client, 8 is a container step of a step (c) (+ c; and fifthly, the processor executes the step ninthly.

Step b, realizing the Session services, the source code segment of the Session services:

the segment of code realizes the container process of the device 8, and the code Server is new Server (); and fourthly, completing the first order, the second order calculation, the third order calculation and the nth order calculation. The SessionServices program pool is completed by the hosting center 4 together with the application server 2.

Sixth, implementation method and device of FIG. 6

FIG. 6 is a front sectional view of step (ii),it is supplementary explanation to fig. 5, that is, the front is the data entry of the application server 2, and mainly realizes the data structure characteristic value: description element (1, 1, 1), description element (2, 2, 2), description element (3, 3, 3), description element (n, n, n.) in front-sideIs the point description element entrance of the three-dimensional coordinate of the description element (1, 1, 1), and the stepIs the point description element entrance of the three-dimensional coordinate of the description element (n, n, n); in the figure, the front is a container 9 data outlet, and the data structure characteristic value is mainly realized: description element (1, 1, 1), description element (2, 2, 2), description element (3, 3, 3), description element (m, m, m.) in frontIs the point description element outlet of the three-dimensional coordinate of the description element (1, 1, 1); step (ii) ofThe point description element exit of the three-dimensional coordinate of the description element (m, m, m), the step (c) and the middle hiding step form a network which is composed of an input layer, a plurality of high layers and an output layer space-time tunnel description element set and is contained in the SessionServices data processing pool.

SessionServices pool data processing code fragment:

seventh, implementation method and device of FIG. 7

FIG. 7 is a description element hiding step of the master and slave containersThe matrix eigenvalues and eigenvector coordinates of (2).

9 is a container of step ninthly, step in the figureStep of projecting description elements of two different time points to two-dimensional x, y projection pointsTo pass throughAfter the step 9, the data stored in the container is a cluster of time + space three-dimensional data elements, and a huge amount of mathematical operation libraries are stored in the device 9 and can be called in real time.

Eighth, method and device for implementing FIG. 8

FIG. 8 is a block diagram of bus servers 4 and (iv), where the app is the application layer; bin/libs virtual machine layer; gust os is the virtual machine operating system layer; 4, server register center, step four is @ WebServices tangent plane interface; 7 is a main container; and 6 is a slave container.

@ WebServices is a tangent enterprise bus interface, integrates @ WebService, and realizes fragments in a source code of a main control center server 4:

the enterprise bus consisting of a plurality of WebServices is integrated through the codes WebServices, so that preparation is made for the matrix operation in the step three.

The @ WebService interface realizes the existing WebService technology. For the client, various WebService client APIs transmit url addresses of wsdl files, the APIs create underlying proxy classes, and the proxies are called to access WebService services. The proxy class changes the method call of the client into request data in the form of the soap and then sends the request data out through the HTTP protocol, and changes the received soap data into a return value to be returned. For a server, the essence of various WebService frameworks is a great Servlet, when a client is remotely called to send request data in a soap format through an http protocol, the client analyzes the data to know which method of which java class needs to be called, then searches or creates the object, calls the method, packages the result returned by the method into data in the soap format, and returns the data in the soap format to the client through an http response message.

WebService is a remote calling technology of cross-programming language and cross-operating system platform, and is a technology that an application program exposes an API (application program interface) which can be called through Web to the outside, namely the application program can be called through the Web by using a programming method. An application that calls this WebService is called a client, and an application that provides this WebService is called a server. From a deep level, WebService is a new platform for establishing an interoperable distributed application program, is a platform, and is a set of standards. It defines how applications achieve interoperability on the Web. XML + XSD, SOAP and WSDL are three major technologies that constitute the WebService platform.

And (3) realizing that a user submits a user to a master control center interface source code segment:

the following description focuses on the functional structure of the bus server 4 and the step (iv).

1. Distributed management of Web service distributed management

Compared with WebServices, WebServices have more mature technology and corresponding standards and management programs in service and service processes; based on the prior art, the WebServices are further innovated to become the WebServices of an enterprise bus, all the advantages of the WebServices are inherited in performance, and the distributed WebServices are integrated to uniformly manage and dispatch. With the increase of the number of the used services and business processes, the management of all the services running in a multi-phase environment, the system performance is very important, a distributed management scheme of an application support environment monitoring management system based on Web services is needed, the remote management problem is solved by SOAP message calling, the management switching of a plurality of servers is solved by a monitoring management tool through a navigation bar tree structure and authority information dynamic switching, the monitoring and management of the Web service container and a workflow engine are achieved through the monitoring and management of the internal information states of the Web service container and the workflow engine, the intuition and richness of the display of monitoring information are ensured through the display of a scalable vector graphic dynamic graphic, the multi-bit integrated skill center management mode is realized, and the main control and auxiliary control functions are completed.

2. Web service choreography Web service orchestration

Web service arrangement of WebServices emphasizes cooperative work and service cooperative capability, and controls interaction of each partial resource through an interaction sequence of messages. The resources participating in interaction are all equal and have no centralized control, so that the aim is achieved, the enterprise bus locking and unlocking technology plays an important role, and according to the prior art, the next step of work can be started after the operation result of one process or thread or workflow is finished; WebServices are different in that the application of the tangent enterprise bus technology solves the important problems of bit synchronization or code element synchronization, group synchronization or frame synchronization, and the WebServices parallels the data of a time axis to a bus tangent plane in the step IV for parallel processing, each process, thread or workflow waits for an operation result, and the next result is immediately executed after the result is obtained, and the processes, threads or workflows are not connected in series in time.

3. Web service organization Web service orchestration

The compiling and arranging are two standard solutions for enterprise connectivity, and the compiling is just like a traffic light and controls the passing of vehicles. The choreography is as if it were a roundabout, without centralized control, but with a series of rules to indicate that vehicles must wait while approaching the intersection until there is room to enter the roundabout system and then find the right time to leave. The visual metaphor is much easier for understanding the Web services organization, the application server 2 forms the data characteristics of Keys, Values and SessionServices after the steps of loading the virtual machine, debugging the virtual machine and explaining the virtual machine, and the characteristics are data models and can be divided into two types: one is a conceptual model, also called an information model, which models data and information from the user's perspective. The method is mainly used for designing the database; the other type is called a data model, which is further divided into model types such as relationship, hierarchy, and mesh, and is established from the viewpoint of a computer system. The programming is that the design flow is carried out on the data model base oriented to the executable scheme: the process compilation uses an executable central process to coordinate internal and external WebServices interactions. The overall target, the related operation and the service calling sequence are controlled through a central flow.

Orchestration and orchestration are two related ways to combine web services to work together. Both of these criteria are important to establish a dynamic, extensible process. Their goal is to provide a range of open, standard protocols to design and execute the interaction of multiple Web services.

4. Web service coordination

The coordination service comprises an activation service, a registration service and a specific coordination protocol service which form a coordination framework, the Web service coordination of WebServices is used for coordinating the transmission of various messages, the service is established according to the coordination protocol and exists in the form of WebService, and a universal coordination framework can be provided in the environment of the WebService so as to solve various applications needing coordination functions. The WebServices has the important characteristic that an open coordination framework is provided, and each specific application can be transmitted to the WebService by the WebServices to design a self-coordination protocol so as to control the whole coordination process and achieve the respective application purpose.

5. Web service transactions

Among distributed systems, the number of subsystems to be subjected to transaction control is not one or two, but N,? WebServices are subjected to separate isolation processing from a container 6 in the transaction processing process, all commands in the transactions are serialized and sequenced, the transactions are not interrupted by command requests sent by other clients in the execution process, each process has a Session mechanism of the process, the sessions are realized by serialization and serialization no matter the sessions are realized by IP, cookies and hidden input, an application program level Session service transaction pool is maintained in a server in a main container 7, submitted transaction objects are placed in the server, and each transaction object is identified by a unique mark ID to form a dictionary object pool.

6. Web service security Web service security

Design choices and implementations that meet the security requirements often adversely affect the performance of the solution. This does not mean that all security techniques used for the solution will result in low performance. Instead, it should be appreciated that Web services solutions that require authentication of business participants, digital signing of message content, and encryption of XML data may have very different performance characteristics depending on the technology and method of protecting the business functions and data exposed by the solution. Three requirements for WebServices security include: (a) authentication, (b) data integrity (dataintegrity), and (c) data confidentiality (data confidentiality).

7. Web service policy Web service policy

WebServices defines a policy as a collection of one or more policy assertions. Web services policies require the provision of a generic model and syntax for describing and communicating with the policies of the Web service, and these assertions specify that traditional requirements and functions (e.g., authentication mode, transport protocol selection) will ultimately be embodied. Some assertions specify requirements and functions that do not embody traditional requirements, but are critical to proper service selection and use, technology specific mechanisms for associating policies with various entities and resources can be freely defined, a unified policy syntax is provided to allow reasoning about both assertions in a consistent manner, and further specifications include the definition of nested assertions to provide more granularity in accounting for the needs of a particular domain. WebServices are enterprise buses, and can correctly judge and participate in WebService, and all actions are finished by WebService.

8. Web service reliable messaging

Each connection between an application and a policy set may select a bus and a messaging engine for the state of the reliable messaging protocol. After the reliable messaging target receives a message from the reliable messaging source, it can deliver the message to the application target at any time, obtaining a different quality of service. These qualities of service range from protecting messages from loss on the network in order to protect the server from failure.

9. quality of service of services

In an enterprise, if the reliability of a system is critical to an organization, the system is a mission critical system of the enterprise to address high level demands, also referred to as quality of service (QoS). For network services, the service quality is improved, that is, the transmission bandwidth is guaranteed, the transmission delay is reduced, and the packet loss rate and the delay jitter of data are reduced. In a broad sense, the quality of service relates to aspects of network applications, and is actually improving the quality of service as long as measures are beneficial to the network applications. In the platform system, WebService is integrated by WebServices; session is integrated with SessionService; the SessionService is integrated by SessionServices; WebServices and Session Services are tightly coupled and loosely coupled matrix operations in a virtualized environment, solving high concurrency problems, security problems, network latency or interruption problems; the key technology of the tangent enterprise bus interface is realized; the key technology of seamless data butt joint of the master container and the slave container is realized; the time and space data structure key technology is realized. A data processing center based on a virtualization master-slave container is a technology created for service quality, provides a better performance for the existing network, and enables various network applications to operate more smoothly.

10. J2EE and. Net

Although the J2EE and NET platforms are commonly used platforms for developing SOA applications, the SOA is not so limited. Platforms like J2EE not only provide a platform for developers to participate in SOA naturally, but also introduce scalability, reliability, availability and performance into the SOA world through their inherent features. A new specification, such as JAXB (JavaAPI for XML binding), for locating XML documents to Java classes; JAXR (Java API for XMLRegostory) is used to specify operations on a UDDI registry (registry); XML-RPC (Java API for XML-based remote Procedure Call) is used in J2EE1.4 to invoke remote services, which makes it easy to develop and deploy Web services that are portable to standard J2EE containers, while at the same time enabling cross-platform (e.g., NET) service interoperability.

11、WS-I Basic Profile

WS-I Basic Profile, provided by Web Services Interoperability Organization (Web Services Interoperability Organization), is a core component required by SOA service test and Interoperability. The service provider may use the Basic Profile test program to test the interoperability of services on different platforms and technologies.

12、SOAP，WSDL，UDDI

WSDL, UDDI and SOAP are the fundamental components of the SOA foundation. WSDL is used to describe services; UDDI is used to register and find services; SOAP, in turn, serves as a transport layer for transporting messages between consumers and service providers. SOAP is the default mechanism for Web services, and other techniques may implement other types of bindings for services. A consumer may look up a service in the UDDI registry (registry), get a WSDL description of the service, and then call the service via SOAP.

Method and device for realizing FIG. 9

Fig. 9 is a functional structure of the bus server 4 and the step iv analyzed in combination with fig. 8, and an enterprise bus is customized by the functional structure of the bus server 4 and the step iv based on the standardization in the prior art, so that the main control center and the auxiliary control center are more reliable, and high-quality data operation is realized.

Ten, 10 implementation method and device

The application server 2 is configured as a virtual machine, which is divided into three main subsystems: firstly, a class loader subsystem; secondly, a running period data area; and thirdly, executing the engine.

Class I loader subsystem

The dynamic class loading functionality of Java is handled by the class loader subsystem. When it first references a class at runtime (not at compile time), it loads, links, and initializes the class file.

Three class loaders, a boot class Loader (boot strap class Loader), an Extension class Loader (Extension class Loader), and an application class Loader (application class Loader), help complete the loading of classes. A boot class loader, responsible for loading a class rt.jar from a boot class path, which is assigned the highest priority; the extension class loader is responsible for loading classes in the ext directory (jre \ lib); the class loader loads class files according to a Delegation Hierarchy Algorithm (deletion Hierarchy Algorithm), which is described later.

The loading process mainly accomplishes three things: obtaining a binary byte stream defining the class through the fully qualified name of the class converts a static storage structure represented by the class byte stream into a runtime data structure of the method area to generate a java.

Initialization: this is the final phase of class loading, where all static variables will be initialized and the static block will be executed. In java, for the initialization phase, there are and only the following five cases that will initialize the requirement class immediately: instantiating an object using the new key, accessing or setting a static field of a class (exception of constant pool already put when modified by final, compiler optimization), calling a class method, all initialize the class where the static field or static method is located; when initializing the class, if the parent class is not initialized, triggering the initialization of the parent class; when the method of java.lang.reflex package is used for reflection calling, if the class is not initialized, the class is initialized first; when the virtual machine is started, a user can initialize a main class (including main) to be executed;

secondly, a running period data area, wherein the running period data area is the working process step of the java data structure parserAs will be explained in detail.

The java data structure analysis process mainly comprises the following four processes: class or interface analysis, field analysis, class method analysis and interface method analysis.

The runtime data area is divided into 5 main components:

(1) the method area (thread sharing) is also stored with the compiled code of the constant static variable JIT (just in time compiler);

(2) a main site for heap memory (thread sharing) garbage recovery;

(3) a position indicator of the bytecode executed by the current thread of the program counter;

(4) java virtual machine stack (stack memory): storing local variables, basic data types and reference variables of objects in a heap memory;

(5) local method stack (C stack): and providing service using a native method for the virtual machine.

Third, execution engine

The bytecode assigned to the run-time data area will be executed by the execution engine. The execution engine reads the bytecode and executes it segment by segment.

The Java virtual machine is a hypothetical computer that can run Java code. Any Java code that is compiled can be guaranteed to run on the system as long as the interpreter is ported to a specific computer according to the virtual machine specification description. The virtual machine implements code fragments as follows:

the source code fragment implements a virtual machine: three class loaders, a boot strap class Loader (boot strap class Loader), an Extension class Loader (Extension class Loader) and an Application class Loader (Application class Loader), help to complete the loading of classes. A boot class loader, responsible for loading a class rt.jar from a boot class path, which loader will be given the highest priority; the extension class loader is responsible for loading classes in the ext directory (jre \ lib); the application program class loader is responsible for loading application program level class paths, and environment variables and the like related to the paths.

Wherein the implementation steps of extensions JDWPJDWP online debugging, namely, the JDWP is used immediately after being debugged; instances community is a local load.

Eleven, method and device for realizing fig. 11

Fig. 11 shows a virtual machine debugging schematic of the application server 2, in which a series of debug extended customized Java debugging tools are developed and customized in the application server 2, through which a debugging command can be sent to the application server 2, and a debugger receives and displays a debugging result. Between the debugger and the debuggee, the debugging command and the debugging result are transmitted through the communication protocol of JDWP. All commands are packaged into JDWP command packets, the JDWP command packets are sent to a debuggee through a transport layer, and after the debuggee receives the JDWP command packets, the commands are analyzed and converted into calls of JVMTI to be run on the debuggee. Similarly, the JVM SI run results are formatted into JDWP packets, sent to the debugger and returned to the JDI call. The debugger developer obtains data through JDI and issues instructions. JDWP is a protocol that acts as a communication bridge between the debugger and the Java virtual machine.

Java debug code fragment in application server 2:

and judging whether the virtual machine is loaded or not, if so, executing debug, and performing read-write operation.

The debug plays the following roles:

1. the check bytecode verifier can check whether the generated bytecode is correct or not, and if the check fails, a check error can be obtained.

2. And (3) file format verification: based on byte stream verification, the byte stream is verified to conform to the specification of the current Class file format and can be processed by the current virtual machine. After the verification is passed, the byte stream will enter the method area of the memory for storage.

3. And (3) metadata verification: based on the storage structure verification of the method area, the byte codes are subjected to semantic verification, and the condition that metadata information which does not conform to java language specifications does not exist is ensured.

4. Byte code verification: based on the storage structure verification of the method area, the data flow and the control flow are analyzed, and the fact that the tested method does not do actions which harm the virtual machine when running is guaranteed.

5. Symbol reference verification: the storage structure verification based on the method area occurs in the analysis stage, and the successful analysis of the symbol reference into the direct reference is ensured, so that the analysis action is ensured to be normally executed. In other words, the matching is checked for information other than the class itself.

6. And outputting the tested parameters.

Twelfth, fig. 12 implementation method and device

FIG. 12 is a diagram of a java data structure interpreter implementation data structure, with myclass representing a private data attribute feature and class1 representing one of the data attribute features. The steps in the figureThe analysis of the virtual machine is indicated by the java data structure interpreter.

The Java data structure interpreter is an important part of a Java virtual machine, and works to convert byte codes into machine codes and run the machine codes on a specific platform. The Java data structure interpreter is equivalent to a "CPU" that runs Java bytecodes, but the "CPU" is not implemented by hardware, but by software. The platform system adopts an SMP + AMP + BMP multi-core CPU processing mode, and the multi-thread program design fully utilizes the processing capacity of the CPU, so that the high-quality high-reliability parallel execution can be realized, and the reasonable performance on the platform system is improved.

The application server 2 is a specification description which is independent of a specific platform and defined by Java byte codes, and is the basis of the independence of the Java platform. The data structure interpreter is responsible for organizing and processing data, and how the data structure interpreter parses the virtual machine looks at the following code segments:

step 1, applying for a free memory to an operating system. The virtual machine applies for the free memory from the operating system as a system, searches the memory allocation table of the virtual machine, then gives the initial address and the end address of the memory segment to the virtual machine, and the virtual machine prepares to load the class file.

And step 2, allocating the memory. The virtual machine allocates memory. The virtual machine obtains memory, divides the heap into memories first, and then allocates the stack memory well.

Step 3, the file is checked and analyzed for class files. If an error is found, an error is returned.

And 4, loading classes. And loading the class. Since no loader is specified, the virtual machine loads all classes in rt.jar to the Method Area of the heap class storage by default using the bootstrap loader, and vm is also loaded to the memory. In the Method Area, reference is made by the notation of the main Method and the runstatic Method, since they are static methods, which are loaded at the time of class loading. Heap is empty and Stack is empty because there are no new objects and threads to execute.

Step 5, executing a main method. Execution starts a thread and starts the main Method, adding the MYCLASS _ CONST constant in the Method Area. Two object objects and a showcase object are arranged in the heap memory, and vm showcase () is executed to realize accumulation until the loading of the subclass is finished; the object is a parent class and a showcase subclass, the virtual machine initializes the parent class first and then initializes the subclasses, and meanwhile, the virtual machine also creates a program counter to indicate a next statement to be executed.

And 6, releasing the memory and finishing the operation.

The code segments carry out parent class initialization through extensions com inheritance com, then sub-classes are initialized, constant parameters of a bus WebService are added into a Method Area, vm showcase is executed to be new vm (), the five steps are realized, and the realization steps are achievedThe purpose of (1).

Thirteen, fig. 13 realizing method and device

FIG. 13 is a diagram of a java data structure interpreter implementing machine instructions, steps of whichThe method is characterized in that a java data structure interpreter realizes a machine code process of a virtual machine, the machine code realization is based on matrix mathematical operation, each matrix eigenvalue is accurate, and the operation of the eigenvalue and the eigenvector is solved.

The virtual machine realizes the machine code process, and the Java code explains five specifications of execution and specific implementation, which are as follows: a virtual machine instruction system; a virtual machine register; a virtual machine stack structure; the virtual machine fragments are recycled; a virtual machine storage area.

The virtual machine passes each bytecode to be executed to the interpreter, translates it into the corresponding machine code, and then executes it by the interpreter. The interpreter reads in a statement each time when executing, and executes a specific operation according to the statement; then the next sentence is read in, and so on. The interpreter includes two main subsystems: one is an expression parser which is responsible for processing the digital expression; the other is an interpreter, which is responsible for the actual execution of the program. When executing code in memory, a data structure called a stack (stack) is used (i.e., the organization and processing of data). The stack is like a container, the object is placed and taken at the same end, and the later the object is placed, the earlier the object is taken out (last-in first-out). By a popular metaphor: the earlier a vehicle parked in the stack is parked at the farther inside, and the vehicle parked later can be driven until the vehicle is driven.

In the application server 2, the application server is completely software-based, and the software-based application server system realizes virtualization; the virtual and software application server is completely embedded into the main control center to form an auxiliary control center, and the auxiliary control center realizes five specifications of a virtual machine: a virtual machine instruction system; a virtual machine register; a virtual machine stack structure; the virtual machine fragments are recycled; a virtual machine storage area. The operation of the main control center and the auxiliary control center is the mathematical operation of the solution of the matrix eigenvalue and the eigenvector through the tight coupling and loose coupling technology.

Fourteen, figure 14 implementing method and device

FIG. 14 is a description meta-space-time coordinate diagram of the application server 2, and 9 is a nine-container in whichProjecting the descriptions of two different time points to a projection point of two-dimensional x and yThe above. Step (ii) ofThe application server 2 realizes the three-dimensional data element conversion, and the operation result is put into a 9-dimensional containerWaiting for bus call after stepAnd step three, coupling and synchronization, the device 9 stores a huge operation library which can be called in real time, and the matrix operation for realizing the main control center and the auxiliary control center is realized through tight coupling and loose coupling, remote calling and matrix operation.

Fifteen, 15 realizing method and device

In fig. 15, 1, 2, 3 and 4 are the same as fig. 3, 6 and 7 are the same as fig. 4, 8 and 9 are the same as fig. 5, and the steps are the third step, the fourth step, the seventh step and the sixth step are the same as fig. 3, and the fifth step is the third stepIs a time point entry element of the three-dimensional coordinates of the description element (1, 1, 1); step (ii) ofIs the time point exit element of the three-dimensional coordinates of the description element (1, 1, 1), stepMethod for preparing a Chinese medicinal compositionAfter the operation of locking and unlocking the bus is carried out in the step (IV), the time spent is the time element; step (ii) ofProjecting two different time point description elements to two-dimensional x, y projection points, and performingThe operation result of (2) is obtained byThrough the step (iv) the enterprise bus is output to the slave container 9, and through the step (iv)And (4) outputting the characteristic values and the characteristic vectors to an outlet of the container 9 from the step (4) and the step (iv), wherein the container 9 is the container in the step (nini), and the characteristic values and the characteristic vectors after the matrix operation are mainly stored.

Devices 1, 2, 3, 4, 6, 7, 8, 9 are connected in the order: the exit of 1 is the entry of 2, the exit of 2 is the entry of 3, the exit of 3 is the entry of 4, the exit of 4 is the step of the SessionServices session cut plane. Step 7 is a @ WebService tangent plane interface, the outlet of the 6 is an inlet of the 7, the outlet of the 7 is an inlet of the 8, and the outlet of the 8 is an outlet stepIs an inlet of 9, and 9 is a step ninthlyAfter the tight coupling and the loose coupling, the remote calling and the matrix operation are carried out, the data binding is output to the application server 2 by the fifth step, and the application client 1 is waited to call; 10 is interceptor, and the step r is big data interception; step (ii) ofIs a database withAnd (6) connecting.Is the application server 2 implementing the steps.

The platform system is divided into a main control center, an auxiliary control center and three-level nodes, wherein the main control center is a main server group (a main container and a slave container); the auxiliary control center is a slave server group (application server); the third-level node is a client application layer; the distributed file server can realize a safe, quick, stable and extensible file storage solution; 1. a configurable distributed architecture is adopted to meet the storage application requirements of different scales; 2. the creative platform adopts an embedding mode based on the three-layer structure, is flexible and convenient to configure, and can break through space-time limitation; 3. the network marketing function realizes a commercial application function; 4. the professional application terminal adopts tunnel virtual connection, so that the communication cost can be reduced.

A virtualized master-slave container based data processing center is configured to include: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, which constitute hardware and software fundamental resources: hardware basic resources mainly include three major types of devices in a network environment, namely: computing (servers), storage (storage devices), and networks (switches, routers, etc.); the software basic resources comprise a stand-alone operating system, middleware, a database and the like.

The virtual bus application server builds a technology to run specialized data, the algorithm and the data are located on different computers, parallel processing is achieved, distributed computing is completely implemented, and the virtual machine can promote data source and machine learning algorithm, class or interface analysis, field analysis, class method analysis and interface method analysis. Such as clustering, regression, classification, graph mode, and optimized access, and provides visual modules. The virtual machine comprises a database for storing and processing any type of data matrix operation, and can process very large matrixes even when the matrixes cannot be written into a memory. Virtual machines provide not only a number of algorithms and tools, but also interfaces with other machine learning and data mining packages. Virtual machine learning and data mining are realized: 1. the integration of big data, the whole platform of the system is divided into a main control center and an auxiliary control center, and the system has obvious advantages in data acquisition, data access, basic architecture, data processing, statistical analysis and data mining, and provides the basic architecture for realizing the integration of the big data; 2. the key technology of big data processing, big data acquisition, big data preprocessing, big data storage and management, big data analysis and mining, big data display and application (big data retrieval, big data visualization, big data application, big data security and the like) are all matrix operations which are established in a virtual environment and are tightly coupled and loosely coupled for remote calling, high-quality data processing algorithms are realized, and a basic framework is provided for realizing the storage, transmission, processing and fusion of mass data of the data through one-dimensional, two-dimensional, three-dimensional, time + space three-dimensional data conversion and processing.

Big data acquisition technology: the various types of structured, semi-structured (or referred to as weak structured) and unstructured massive data obtained by the data matrixing mode are the basis of a big data model. The mathematical operation of a matrix eigenvalue and eigenvector solving method of the big data model is realized, and the big data collecting technologies such as distributed high-speed high-reliability data crawling or acquisition, high-speed data full mapping and the like are broken through in a key way; breaking through big data integration technologies such as high-speed data analysis, conversion and loading; and designing a quality evaluation model and developing a data quality technology.

Big data acquisition is generally divided into: 1. the intelligent perception layer: the system mainly comprises a data sensing system, a network communication system, a sensing adaptation system, an intelligent identification system and a software and hardware resource access system, and realizes intelligent identification, positioning, tracking, access, transmission, signal conversion, monitoring, primary processing and management and the like of structured, semi-structured and unstructured mass data. Technologies such as intelligent identification, perception, adaptation, transmission, access and the like aiming at a big data source are realized through a data model, and a third-level node of the platform realizes the function; 2. a basic supporting layer: the system platform auxiliary control center provides basic support environments such as a virtual server, a debugger and a data structure interpreter which are needed by a big data service platform, wherein the basic support environments include a database of structured, semi-structured and unstructured data, internet of things resources and the like, so that a distributed virtual storage technology, a visual interface technology of big data acquisition, storage, organization, analysis and decision operation, a network transmission and compression technology of big data, a big data privacy protection technology and the like are realized.

Big data preprocessing technology: the operations of analyzing, extracting and cleaning the received data are mainly completed by the matrix eigenvalue and the eigenvector.

1) Extracting: because the acquired data may have various structures and types, the data extraction process can help us convert the complex data into a single or convenient-to-process configuration so as to achieve the purpose of rapid analysis and processing.

2) Cleaning: for large data, it is not all valuable, some data are not the content we are interested in, and other data are the interference terms of complete errors, so the data are filtered and "denoised" to extract the valid data.

Big data storage and management technology: through one-dimensional, two-dimensional, three-dimensional, time + space three-dimensional data conversion and processing, a basic framework is provided for realizing storage, transmission, processing and fusion of mass data of the data. The big data storage and management uses a memory to store the collected data, establish a corresponding database, and manage and call the database. The method mainly solves the management and processing technology of complex structured, semi-structured and unstructured big data. The method mainly solves several key problems of large data such as storage, representation, processing, reliability and effective transmission. Developing a reliable distributed file system, energy efficiency optimized storage, calculation integration storage, redundancy removal of big data and a big data storage technology with high efficiency and low cost; the distributed non-relational big data management and processing technology, the data fusion technology of heterogeneous data, the data organization technology and the research big data modeling technology are broken through; breaking through big data index technology; the technologies of big data movement, backup, copy and the like are broken through; and developing big data visualization technology.

Big data analysis and mining technology: namely, the analysis and the mining of the data structure characteristics are carried out by utilizing a matrix operation library. Improving existing data mining and machine learning techniques; developing novel data mining technologies such as data network mining, specific group mining, graph mining and the like; the big data fusion technology based on object data connection, similarity connection and the like is broken through; the large data mining technology facing the field, such as user interest analysis, network behavior analysis, emotion semantic analysis and the like, is broken through.

Data mining is the process of extracting information and knowledge hidden in it that is not known a priori but is potentially useful from a large, incomplete, noisy, fuzzy, random, real-world data. The technical methods involved in data mining are many and there are many classifications. According to the mining task, the classification or prediction model discovery, data summarization, clustering, association rule discovery, sequence pattern discovery, dependency or dependency model discovery, abnormity and trend discovery and the like can be divided; according to the mining object, the mining object can be divided into a relational database, an object-oriented database, a spatial database, a temporal database, a text data source, a multimedia database, a heterogeneous database, a legacy database and a world wide Web (Web); according to the excavation method, the method can be roughly divided into: a machine learning method, a statistical method, a neural network method, and a database method. In machine learning, the following steps can be performed: inductive learning methods (decision trees, rule induction, etc.), case-based learning, genetic algorithms, etc. The statistical method can be subdivided into: regression analysis (multiple regression, autoregressive, etc.), discriminant analysis (bayes discriminant, fisher discriminant, nonparametric discriminant, etc.), cluster analysis (systematic clustering, dynamic clustering, etc.), heuristic analysis (principal component analysis, correlation analysis, etc.), and the like. The neural network method can be subdivided into: forward neural networks (BP algorithm, etc.), ad hoc neural networks (ad hoc feature mapping, competitive learning, etc.), and the like. The database method is mainly a multidimensional data analysis or OLAP method, and in addition, an attribute-oriented induction method is also provided.

From the perspective of the excavation task and the excavation method, the method is mainly characterized in that: 1. and (4) performing visual analysis. Data visualization is the most basic function, whether for the average user or the expert in data analysis. The data imaging can make the data speak by itself, and the user can intuitively feel the result. 2. And (3) a data mining algorithm. Imaging is the translation of machine language to human, while data mining is the native language of a machine. The analysis of segmentation, clustering and isolated points also has various algorithms of various types of five-flower eight doors, so that the data are refined and the value is mined. These algorithms must be able to cope with large data volumes while also having a high processing speed. 3. And (4) predictive analysis. Predictive analysis may allow the analyst to make some prospective decisions based on the results of imaging analysis and data mining. 4. And a semantic engine. The semantic engine needs to be designed with enough artificial intelligence to actively extract information from the data. The language processing technology comprises machine translation, emotion analysis, public opinion analysis, intelligent input, question and answer system and the like. 5. Data quality and data management. Data quality and management are the best practices for management, and the data processing through standardized processes and machines can ensure that a preset quality of analysis results and consistent data representation are obtained. For embedded specialized data processing techniques based on virtualization, everything is a matrix, e.g., several matrices can be combined into one variable, e.g., a time series. These matrices may be accessed one by one or as a single large matrix. Several variables may be combined into one sample, e.g., a sample with input values and target values in the classification. Many samples may form a data set that may be subject to storage or splitting operations in a cross-validation test. The data set may be accessed sample by sample, or may be accessed using a large matrix as an input value and a large matrix as a sample of a target value.

Based on the above four data structure parsing types, it is how the application server 2 can facilitate learning algorithms for data sources and machines, and each data has certain characteristics, such as: the class has the characteristic of privacy, namely the class which has the characteristic of class1, class2, class3 and class attribute, and each class characteristic attribute is translated and organized into a machine language in a key and value mode to form a characteristic value and a characteristic vector of a machine code. And the application server 2 provides a vector space to realize the eigenvalue decomposition.

Compared with the prior art, the invention discloses a data processing center based on a virtualization master-slave container, and the key technology is as follows:

1. through the combination of the WebService technology and the SOAP technology, the self WebServices and Sessionserves bus technologies are created, and scientific calculation and high-performance communication are realized. The technical features are particularly applicable to optical communication networks.

2. The matrix computing technology provides a basic framework for realizing storage, transmission, processing and fusion of mass data of data through one-dimensional, two-dimensional, three-dimensional, time + space three-dimensional data conversion and processing.

3. The data processing technology with superior performance of the master control center can realize hot backup and can quickly recover when a fault occurs, so that the cloud computing application is more stable and safer; the communication is smoother due to the huge virtual space, all user data are stored in the container, and the user spaces are logically isolated from each other, so that a reliable private cloud space is formed.

4. The service quality is reliable. The private cloud is inside the enterprise instead of a certain remote data center, so when the application based on the private cloud is accessed by the staff of the company, the service quality of the application based on the private cloud is very stable, and the application based on the private cloud is not affected by the accidental exception of a remote network.

5. The technology of the complete software application service server completely embeds professional application data into the main control center to form an auxiliary control center, realizes access of unlimited application servers, greatly saves cost, and can put more saved cost into the main control center and a route.

6. The adjustable communication technology can realize the self-adaptability of internal data exchange and high concurrent data throughput, and the application server can be exerted to the optimal performance.

7. The data processing of the main control center and the auxiliary control center are organically coupled, so that remote calling is realized, and the optimal performance of the whole cloud platform is exerted.

8. And the data structure is subjected to matrixing data processing, so that the quality of data input from the application server to the master control center is more reliable.

9. The personalized customization function can be realized through software, so that the whole application platform system can play different functions in different fields.

10. The virtual bus application server builds a technology to run specialized data, algorithms and data are located on different computers, parallel processing is achieved, distributed computing is completely implemented, the virtual bus application server is a model technology for large data application and is particularly suitable for application of the Internet of things.

The data processing center based on the virtualization master-slave container has the main purposes that: by implementing the data processing center based on the virtualization master-slave container, an elastic and safe virtualization cluster is designed and deployed. The cluster logically isolates the cloud computing user spaces by using a virtualization technology, and effectively solves the contradiction between the sharing performance and the safety. The cluster also has good expandability, high availability and lower cost. Can be applied to the fields: large 3D digital authoring; domestic backbone

Finally, it will be understood by those skilled in the art that the above-described invention is not limited to the above-described embodiments, and that the above-described embodiments and descriptions are only illustrative of the principles of the invention, and that various changes and modifications may be made without departing from the spirit and scope of the invention, which fall within the scope of the invention as claimed.

Claims

1. A virtualized master-slave container based data processing center configured to include: (1, 2, 3, 4, 5, 6, 7, 8, 9, 10) and capable of performing the steps of: (r, c) data processing method in which a computer program (instructions) is stored, characterized in that the program (instructions), when executed by a processor, implements (r,) And (5) carrying out the following steps.

2. The data processing center based on the virtualized master-slave container as in claim 1, wherein step (r) is performed to implement user data user.

3. The data processing center based on the virtualized master-slave container as recited in claim 1, wherein step (c) is performed to implement user data user.

4. The data processing center based on the virtualized master-slave container as claimed in claim 1, wherein step (c) is performed to realize tight coupling and loose coupling, remote invocation, and matrix operation.

5. The data processing center based on the virtualized master-slave container as claimed in claim 1, wherein the step (iv) is performed to perform the @ WebServices enterprise bus interface, perform deep integration on the @ WebService, and implement high-quality operations.

6. A virtualized master-slave container based data processing center according to claim 1 characterised in that step (v) is executed to receive the data of the container (9) and output the data bound to the application server (2).

7. The data processing center based on the virtualized master-slave container as claimed in claim 1, wherein the step (c) of executing the computer program (instruction) by the processor is to perform @ WebService interface, realize slave container interface, and realize data synchronization and diary synchronization between the master container and the slave container, and when performing high quality operation, is responsible for integrating the user data into the master container.

8. A virtualized master-slave container based data processing center as in claim 1 where the step of (the) executing the computer program (instructions) by the processor(s) performs SessionServices integration for SessionServices to achieve three dimensional data pool SessionServices data processing.

9. The virtual master-slave container-based data processing center of claim 1, wherein the step (nini) of executing the computer program (instructions) by the processor implements data element eigenvalue and eigenvector operation and eigenvalue decomposition.

10. A virtualized master-slave container based data processing center as in claim 1 where the step (in r) of executing the computer program (instructions) by the processor implements big data interception.

11. A virtualized master-slave container based data processing center as in claim 1 where the steps of executing computer programs (instructions) by the processor(s) (ii)) And realizing database connection.

12. A virtualized master-slave container based data processing center as in claim 1 where the steps of executing computer programs (instructions) by the processor(s) (ii)) And realizing the operation of the characteristic value and the characteristic vector of the data structure element.

13. A virtualized master-slave container based data processing center as in claim 1 where the steps of executing computer programs (instructions) by the processor(s) (ii)) And loading the virtual machine and the user application program.

14. A substrate according to claim 1In a data processing center that virtualizes a master-slave container, the steps of executing computer programs (instructions) by a processor(s) (ii)) And realizing debugging and testing of the virtual machine: the check byte code checker checks whether the generated byte code is correct; verifying the file format, namely verifying that the byte stream conforms to the specification of the current Class file format and can be processed by the current virtual machine based on byte stream verification; metadata verification, namely performing semantic verification on the byte codes based on the storage structure verification of the method area to ensure that metadata information which does not conform to java language specifications does not exist; byte code verification, namely verifying a storage structure based on a method area, and ensuring that the tested method does not do actions harmful to the virtual machine when running through analyzing data flow and control flow; and symbol reference verification and method area-based storage structure verification.

15. A virtualized master-slave container based data processing center as in claim 1 where the steps of executing computer programs (instructions) by the processor(s) (ii)) And the data structure analysis is realized, and the bytecode is converted into a machine code and operated in a specific device.

16. A virtualized master-slave container based data processing center as in claim 1 where the steps of executing computer programs (instructions) by the processor(s) (ii)) The data structure feature analysis is realized, machine translation is carried out and arranged into machine language, the feature value and the feature vector of the machine code are formed, and the application server (2) provides a vector space to realize the feature value decomposition.

17. According to claimA virtualized master-slave container based data processing center as in claim 1 wherein the steps of executing computer programs (instructions) by the processor(s) (ii) are) The tight coupling and loose coupling, remote calling and matrix operation of the application server (2) and the container are realized, and data synchronization is carried out with the step (III) to realize data synchronization of the main control center and the auxiliary control center.

18. A virtualized master-slave container based data processing center configured to include: (1, 2, 3, 4, 5, 6, 7, 8, 9, 10) device architecture, the connection order of which: (1) is the inlet of (2), (2) is the inlet of (3), (4), (3), (4) is the inlet of (8), (8) is the inlet of (9), and the steps can be performed: (r), (c), c) data processing system having stored thereon a computer program (instructions), characterized in that the program (instructions), when executed by a processor, effects (r), (c, ) And (5) carrying out the following steps.

19. A virtualized master-slave container based data processing centre according to claim 18, characterised in that means (1) are provided for implementing step (r).

20. A virtualized master-slave container based data processing centre according to claim 18, characterised in that means (2) are provided for carrying out step(s).

21. A virtualized master-slave container based data processing centre according to claim 18, characterised in that means (3) are provided for carrying out step (c).

22. A virtualized master-slave container based data processing centre according to claim 18, characterised in that means (4) are provided for implementing step (r).

23. A data processing centre based on virtualized master-slave containers according to claim 18, where means (5) are provided for carrying out the steps (c, r, )。

24. a virtualized master-slave container based data processing centre according to claim 18, characterised in means (6) for implementing step (c).

25. The data processing center based on virtualized master-slave container according to claim 18 or 24, characterized in that the device (7) comprises means (4, 8, 6, 9) for implementing the steps (c), (.

26. A virtualized master-slave container based data processing centre according to claim 18 or 25, characterised in that means (8) are provided for carrying out step (b).

27. A virtualized master-slave container based data processing centre according to claim 18 or 25, c h a r a c t e r i z e d i n that means (9) are arranged to carry out the steps (ninu,)。

28. a virtualized master-slave container based data processing centre according to claim 18, characterised in that means (10) are arranged to implement step (r).

29. A virtualized master-slave container based data processing centre according to claim 18 or 20, characterised in that means (2) are provided for carrying out the steps (a), (b), (c), (d), ()。