Distributed databases are decentralized
Publish: 2021-04-01 05:35:08
1. Sharing from regional TV network: in a system with many nodes, each node has a high degree of autonomy. Nodes can connect freely to each other to form a new connection unit. Any node may become the stage center, but it does not have the mandatory central control function. The influence between nodes will form nonlinear causality through the network. This open, flat and equal system phenomenon or structure is called & quot; Decentralization & quot;.
2. Not only distributed databases are relational, but now mainstream databases are relational
and mesh database are rare.
and mesh database are rare.
3. Distributed
means cluster
means many computers work together. The advantage is that the ordinary computer can be combined into a distributed system without buying special server equipment. The disadvantage is that there is a certain amount of bandwidth that will be lost in the interaction between internal machines
means cluster
means many computers work together. The advantage is that the ordinary computer can be combined into a distributed system without buying special server equipment. The disadvantage is that there is a certain amount of bandwidth that will be lost in the interaction between internal machines
4. 1. Distributed database is a kind of database, which is the combination of database technology and network technology< Each has its own advantages and disadvantages. Distributed database can be divided into two types: logical partition, physical distribution and logical distribution, and physical centralization
Yes, distributed data files are convenient for database management and maintenance
distributed database systems usually use smaller computer systems, each computer can be placed in a separate place, each computer has a complete of the DBMS, and has its own local database. Many computers located in different places are interconnected through the network to form a complete and global large database< This method overcomes the weakness of the physical center database organization
1. First of all, the cost of data transmission is reced, because most of the access operations to the database are for local databases, rather than for databases in other locations
2. Secondly, the reliability of the system is greatly improved, because when the network fails, the operation of the local database is still allowed, and the failure of one location does not affect the processing work of other locations. Only when the data of the failure location is accessed, it will be affected to some extent< It is easy to expand the system, add a new local database, or expand an appropriate small computer in a certain location. However, some functions have to pay a higher price
for example, in order to deploy activities in several locations, the performance of transaction management is more expensive than that in a central database, and even offsets many other advantages< The main characteristics of distributed database system are as follows:
1
2. Computers all over the country are connected by data communication networks
3. Overcome the weakness of central database: rece the cost of data transmission
4. The reliability of the system is improved. If the local system fails, other parts can continue to work
5. The location of each database is transparent to facilitate the expansion of the system
6. In order to coordinate the transaction activities of the whole system, the performance cost of transaction management is high
data fragmentation
type:
(1) horizontal fragmentation: according to certain conditions, all tuples of the global relationship are divided into several disjoint subsets, and each subset is a fragment of the relationship
(2) vertical slicing: the attribute set of a global relationship is divided into several subsets, and projection operations are performed on these subsets. Each projection is called vertical slicing
(3) export slicing: also known as export horizontal slicing, that is, the condition of horizontal slicing is not the condition of this relationship attribute, but the condition of other relationship attributes
(4) mixed slicing: the combination of the above three methods. They can be divided horizontally and then vertically, or vertically and then horizontally, or in other forms, but their results are different
condition:
(1) completeness condition: all data of global relationship must be mapped to fragments, and data belonging to global relationship is not allowed to belong to any fragment
(2) reconfigurable condition: it must be ensured that each segment of the same global relationship can reconstruct the global relationship. For horizontal slicing, global relations can be reconstructed by parallel operations; For vertical fragmentation, join operations can be used to reconstruct the global relationship
(3) disjoint condition: it requires that all data fragments obtained after a global relationship is segmented do not overlap each other (except for the primary key of vertical partition)< (1) centralized: all data segments are arranged in the same site
(2) segmentation: all data has only one , which is divided into several logical segments, each of which is assigned to a specific site
(3) full replication: data is stored repeatedly in each site. That is, there is a complete of the data on each site
(4) hybrid: This is a distribution method between split and full < At present, more and more distributed database allocation algorithms are used to find the optimal solution, such as genetic algorithm, annealing mechanism and so on
Yes, distributed data files are convenient for database management and maintenance
distributed database systems usually use smaller computer systems, each computer can be placed in a separate place, each computer has a complete of the DBMS, and has its own local database. Many computers located in different places are interconnected through the network to form a complete and global large database< This method overcomes the weakness of the physical center database organization
1. First of all, the cost of data transmission is reced, because most of the access operations to the database are for local databases, rather than for databases in other locations
2. Secondly, the reliability of the system is greatly improved, because when the network fails, the operation of the local database is still allowed, and the failure of one location does not affect the processing work of other locations. Only when the data of the failure location is accessed, it will be affected to some extent< It is easy to expand the system, add a new local database, or expand an appropriate small computer in a certain location. However, some functions have to pay a higher price
for example, in order to deploy activities in several locations, the performance of transaction management is more expensive than that in a central database, and even offsets many other advantages< The main characteristics of distributed database system are as follows:
1
2. Computers all over the country are connected by data communication networks
3. Overcome the weakness of central database: rece the cost of data transmission
4. The reliability of the system is improved. If the local system fails, other parts can continue to work
5. The location of each database is transparent to facilitate the expansion of the system
6. In order to coordinate the transaction activities of the whole system, the performance cost of transaction management is high
data fragmentation
type:
(1) horizontal fragmentation: according to certain conditions, all tuples of the global relationship are divided into several disjoint subsets, and each subset is a fragment of the relationship
(2) vertical slicing: the attribute set of a global relationship is divided into several subsets, and projection operations are performed on these subsets. Each projection is called vertical slicing
(3) export slicing: also known as export horizontal slicing, that is, the condition of horizontal slicing is not the condition of this relationship attribute, but the condition of other relationship attributes
(4) mixed slicing: the combination of the above three methods. They can be divided horizontally and then vertically, or vertically and then horizontally, or in other forms, but their results are different
condition:
(1) completeness condition: all data of global relationship must be mapped to fragments, and data belonging to global relationship is not allowed to belong to any fragment
(2) reconfigurable condition: it must be ensured that each segment of the same global relationship can reconstruct the global relationship. For horizontal slicing, global relations can be reconstructed by parallel operations; For vertical fragmentation, join operations can be used to reconstruct the global relationship
(3) disjoint condition: it requires that all data fragments obtained after a global relationship is segmented do not overlap each other (except for the primary key of vertical partition)< (1) centralized: all data segments are arranged in the same site
(2) segmentation: all data has only one , which is divided into several logical segments, each of which is assigned to a specific site
(3) full replication: data is stored repeatedly in each site. That is, there is a complete of the data on each site
(4) hybrid: This is a distribution method between split and full < At present, more and more distributed database allocation algorithms are used to find the optimal solution, such as genetic algorithm, annealing mechanism and so on
5. A distributed database is a single logical database in front of users, but it is actually composed of a group of databases stored on multiple computers. The databases on several computers can be modified and accessed simultaneously through the network, and each database is controlled by its local DBMS. In distributed database, each database server cooperatively maintains the consistency of global database. Each computer in the system is called a node. If a node has database management software, it is called database server. If a node is an application that requests information from the server, it is called a client. In Oracle client, execute database application, access data information and interact with users. In the server, Oracle Software is executed to deal with concurrent and shared data access to Oracle database. Oracle allows the above two parts to be on the same computer, but it is more effective when the client part and the server part are on different computers connected by the network. Distributed processing is a process in which multiple processors share a single task. An example of distributed processing in Oracle database system is that the client and server are located on different computers connected to the network. There are multiple processors on a single computer, and different processors execute client applications separately. SQL * net is Oracle network interface, which allows to run on Oracle tools and servers of network workstations, and access, modify, share and store data on other servers. Saql * net can be regarded as the program interface of network communication. SQL * Net provides a distributed database and distributed processing for oarcle by using communication protocol and application program interface (API). SQL * net driver provides an interface between Oracle process running on database server and user process of Oracle tool. Each server participating in the distributed database manages the database independently, as if each database is not a networked database. Each database is managed independently, which is called site autonomy. Site autonomy has the following advantages: the nodes of the system can reflect the logical organization of the company. The local data is controlled by the local database administrator, so that each database administrator's responsibility domain is smaller and can be better managed. As long as one database and network are available, the global database can be partially available. It will not stop all operations or cause performance bottleneck e to a database failure. Fault recovery is usually carried out on a single node. Each local database has a data dictionary. The node can upgrade the software independently. Schema objects can be accessed from all nodes of distributed database, so just like non distributed local DBMS, a mechanism must be provided to reference an object in local database. Distributed DBMS must provide a naming mode so that an object in the distributed database can be uniquely identified and referenced in the application. In general, uniqueness is implemented at each level of the hierarchy. Distributed DVMS simply extends the hierarchical naming model and implements the unique database naming on the network. Therefore, the global object name of an object is guaranteed to be unique in the distributed database. Oracle allows the use of Wa object names in SQL statements to refer to schema objects (tables, views, and proceres) in distributed databases. In Oracle, the global name of a schema object consists of three parts: the schema name containing the object, the object name, and the database name. Its form is as follows: Scott [email protected] Where Scott is the schema name, EMP is the table name, and the @ symbol is followed by the database name. A remote query is a query that selects information from one or more remote tables, These tables reside in the same remote node. A distributed query can retrieve data from two or more nodes. A distributed update can modify data of two or more nodes. A remote transaction is a transaction, which contains one person or more remote statements. All references are on the same remote node, It contains one or more statements to modify the data of two or more different nodes in a distributed database, The two-stage commit mechanism ensures that all database servers participating in distributed transactions commit or roll back all the statements in the transaction. Oracle distributed database system architecture can provide location transparency for end users and applications by Oracle database administrator Procere provides location transparency in Oracle distributed database system. Oracle allows remote data to be referenced in select, insert, update, delete, select... For update and lock table statements. Query includes join, aggregate, subquery and select... For update. Local and remote tables and views can be referenced. Local and remote tables can be referenced for update, insert, delete, and lock table statements. Note that when referencing long and long raw columns, sequences, modified tables and blocked tables, they must be in the same node. Oracle does not allow remote DDL statements. In a single site or distributed database, all transactions are terminated with commit or rollback statements. Oracle provides two mechanisms to realize the transparency of table plication in distributed database: table snapshot provides asynchronous table plication; The trigger implements the plication of synchronized tables. In both cases, the transparency of table plication is realized.
6.
1. Distributed database is a kind of database, which is the combination of database technology and network technology
Each has its own advantages and disadvantages. Distributed database can be divided into two types: logical partition, physical distribution and logical distribution, and physical centralizationYes, distributed data files are convenient for database management and maintenance
7.
The distributed database supported by distributed computer system is the combination of database technology and computer network technology. Compared with the common centralized database, although the data of distributed database is stored in various points of the network, it can be shared by all users on the network. Legitimate users anywhere can easily obtain and process the required data, just as the data is on their local computer
8. Distributed software systems is a software system that supports distributed processing and performs tasks on a multiprocessor architecture interconnected by communication networks. It includes distributed operating system, distributed programming language and its compiler (interpreter) system, distributed file system and distributed database system
distributed operating system is responsible for managing distributed processing system resources and controlling the running of distributed programs. It differs from the centralized operating system in resource management, process communication and system architecture
distributed programming language is used to write distributed programs running on distributed computer systems. A distributed program is composed of several program moles that can be executed independently. They are distributed on multiple computers of a distributed processing system and executed simultaneously. Compared with the centralized programming language, it has three characteristics: distribution, communication and robustness
distributed file system has the ability of remote file access and transparently manages and accesses the files distributed on the network
distributed database system is composed of several database systems distributed on multiple computer nodes, which provides effective access means to manipulate the sub databases on these nodes. Distributed database can be regarded as a complete database in use, but in fact it is distributed in geographically dispersed nodes. Of course, the sub databases distributed on each node are logically related< Distributed database system is composed of several stations. These stations are also called nodes. They are connected together in the communication network. Each node is an independent database system. They have their own database, central processing unit, terminal and local database management system. Therefore, the distributed database system can be regarded as a combination of a series of centralized database systems. They belong to the same system in logic, but they are distributed in physical structure<
distributed database system has become an important field of information processing and is developing rapidly for the following reasons:
1. It can solve the problem that the organization is scattered and the data needs to be connected with each other. For example, in the banking system, the head office and branches are located in different cities or regions in the city. In business, they need to process their own data, and also need to exchange and process with each other, which requires a distributed system
2. If an organization needs to add new relatively independent organizational units to expand the organization, the distributed database system can be expanded with the least impact on the current organization
3. The need of load balancing. The decomposition of data maximizes the local application, which minimizes the interference between processors. The critical bottleneck can be avoided by sharing the load among processors
4. When there are several database systems in the existing organization, and the necessity of global application increases, the distributed database system can be constructed from these databases from bottom to top
5. The probability of failure of distributed database system of the same scale will not be lower than that of centralized database system, but because the impact of failure is limited to local data application, the reliability of the whole system is relatively high
characteristics
1. In the distributed database system, the concept of centralized control is not emphasized. It has a hierarchical control structure based on the global database administrator, but each local database administrator has a high degree of autonomy
2. In the distributed database system, the concept of data independence is also important, but a new concept is added, which is distributed transparency. The so-called distributed transparency is that when writing a program, it seems that the data is not distributed, so transferring the data will not affect the correctness of the program. But the execution speed of the program will be reced
3. Different from the centralized database system, data rendancy is regarded as a required feature in the distributed system. The reason lies in: firstly, if the data is copied in the required nodes, the local applicability can be improved. Secondly, when a node fails, it can operate the replicated data on other nodes, so it can increase the effectiveness of the system. Of course, the evaluation of optimal rendancy in distributed system is very complex
the types of distributed systems can be roughly classified into three categories:
1. Distributed data, but there is only one total? According to the database, there is no local database< Each layer has its own database
3. Fully decentralized distributed network, without central control part, there can be a variety of connection modes between nodes, such as loose connection, tight connection, dynamic connection, broadcast notification connection, etc.
distributed operating system is responsible for managing distributed processing system resources and controlling the running of distributed programs. It differs from the centralized operating system in resource management, process communication and system architecture
distributed programming language is used to write distributed programs running on distributed computer systems. A distributed program is composed of several program moles that can be executed independently. They are distributed on multiple computers of a distributed processing system and executed simultaneously. Compared with the centralized programming language, it has three characteristics: distribution, communication and robustness
distributed file system has the ability of remote file access and transparently manages and accesses the files distributed on the network
distributed database system is composed of several database systems distributed on multiple computer nodes, which provides effective access means to manipulate the sub databases on these nodes. Distributed database can be regarded as a complete database in use, but in fact it is distributed in geographically dispersed nodes. Of course, the sub databases distributed on each node are logically related< Distributed database system is composed of several stations. These stations are also called nodes. They are connected together in the communication network. Each node is an independent database system. They have their own database, central processing unit, terminal and local database management system. Therefore, the distributed database system can be regarded as a combination of a series of centralized database systems. They belong to the same system in logic, but they are distributed in physical structure<
distributed database system has become an important field of information processing and is developing rapidly for the following reasons:
1. It can solve the problem that the organization is scattered and the data needs to be connected with each other. For example, in the banking system, the head office and branches are located in different cities or regions in the city. In business, they need to process their own data, and also need to exchange and process with each other, which requires a distributed system
2. If an organization needs to add new relatively independent organizational units to expand the organization, the distributed database system can be expanded with the least impact on the current organization
3. The need of load balancing. The decomposition of data maximizes the local application, which minimizes the interference between processors. The critical bottleneck can be avoided by sharing the load among processors
4. When there are several database systems in the existing organization, and the necessity of global application increases, the distributed database system can be constructed from these databases from bottom to top
5. The probability of failure of distributed database system of the same scale will not be lower than that of centralized database system, but because the impact of failure is limited to local data application, the reliability of the whole system is relatively high
characteristics
1. In the distributed database system, the concept of centralized control is not emphasized. It has a hierarchical control structure based on the global database administrator, but each local database administrator has a high degree of autonomy
2. In the distributed database system, the concept of data independence is also important, but a new concept is added, which is distributed transparency. The so-called distributed transparency is that when writing a program, it seems that the data is not distributed, so transferring the data will not affect the correctness of the program. But the execution speed of the program will be reced
3. Different from the centralized database system, data rendancy is regarded as a required feature in the distributed system. The reason lies in: firstly, if the data is copied in the required nodes, the local applicability can be improved. Secondly, when a node fails, it can operate the replicated data on other nodes, so it can increase the effectiveness of the system. Of course, the evaluation of optimal rendancy in distributed system is very complex
the types of distributed systems can be roughly classified into three categories:
1. Distributed data, but there is only one total? According to the database, there is no local database< Each layer has its own database
3. Fully decentralized distributed network, without central control part, there can be a variety of connection modes between nodes, such as loose connection, tight connection, dynamic connection, broadcast notification connection, etc.
Hot content