Welcome to my “Modern database management system.”
PostgreSQL is highly compatible with Oracle DB and is written for existing commercial DBMS and can be applied immediately without application modification. In particular, it can save about 80% of the cost compared to the performance of Oracle DBMS and it is compatible with Amazon Web Service (AWS) DBMS 'Aurora' at the end of last year and it is becoming popular.
According to the DBMS rankings database engine (DB-engines) on September 9, PostgreSQL ranked 4th after Oracle, MySQL, and MS SQL as of August 2017.
The DB engine quantifies the monthly frequency of technical discussions of IT related Q & A sites such as keyword search in search engines such as Google and Bing, Google Trend, stack overflow, DBA stack exchange, related job search and job search, and social network (Twitter) Announces monthly rankings.
Oracle, MySQL, and MS SQL still scored between 1200 and 1300, according to the report. The third place, MS SQL, has 1225.47 points, but PostgreSQL, the fourth place, has a big difference of 369.76 points. However, compared to August last year, Oracle was down 59.85, MySQL was down 16.73, and PostgreSQL was up 54.51. Mongolia DB ranked fourth, and IBM DB2 ranked fifth.
According to the DBMS rankings database engine (DB-engines) on September 9, PostgreSQL ranked 4th after Oracle, MySQL, and MS SQL as of August 2017.
The DB engine quantifies the monthly frequency of technical discussions of IT related Q & A sites such as keyword search in search engines such as Google and Bing, Google Trend, stack overflow, DBA stack exchange, related job search and job search, and social network (Twitter) Announces monthly rankings.
Oracle, MySQL, and MS SQL still scored between 1200 and 1300, according to the report. The third place, MS SQL, has 1225.47 points, but PostgreSQL, the fourth place, has a big difference of 369.76 points. However, compared to August last year, Oracle was down 59.85, MySQL was down 16.73, and PostgreSQL was up 54.51. Mongolia DB ranked fourth, and IBM DB2 ranked fifth.
The cloud-based DBMS is also gradually taking its place. Amazon Dynamo DB, Microsoft Azure SQL database, and Google Big Query also ranked higher than last year.
Domestic DBMSs are also noticeable. Tmax of Tmaxsoft ranked 113th, up 0.41 points compared to August last year, and open source DBMS Cub ride was ranked 115th in the same period last year, compared to 122nd in the same period last year. Altibase also ranked 127th.
On the other hand, open source DBMS middle-line My SQL, PostgreSQL, Mongolian DB, Cassandra, and Ladies were the most popular. Also, the popularity of the time series DBMS in the middle of the DBMS type has been increasing.
A time series DBMS is used for storing and analyzing log data generated continuously from various IT infrastructures such as Internet (IoT) equipment, sensors, and applications in real time. Recently, it is analyzed that the application range is increasing as a large amount of real-time IoT and big data processing demands are increased at a low cost.
Domestic DBMSs are also noticeable. Tmax of Tmaxsoft ranked 113th, up 0.41 points compared to August last year, and open source DBMS Cub ride was ranked 115th in the same period last year, compared to 122nd in the same period last year. Altibase also ranked 127th.
On the other hand, open source DBMS middle-line My SQL, PostgreSQL, Mongolian DB, Cassandra, and Ladies were the most popular. Also, the popularity of the time series DBMS in the middle of the DBMS type has been increasing.
A time series DBMS is used for storing and analyzing log data generated continuously from various IT infrastructures such as Internet (IoT) equipment, sensors, and applications in real time. Recently, it is analyzed that the application range is increasing as a large amount of real-time IoT and big data processing demands are increased at a low cost.
The database (English: database, DB) is a collection of organized data [1] In other words, it is a bundle of public data that can be stored and operated by integrated information of various application systems as a created list.
As technology advances in processors, computer memory, computer storage, and computer networks, the size, functionality, and performance of databases and DBMSs have increased in order of rank. The evolution of database technology is divided into three periods, depending on the data model or structure: an intrusion database, [2] a SQL / relational database, and a relational database.
The relational model originally proposed by Edgar F. Cudd in 1970 began with this tradition, arguing that applications should retrieve data based on their content, not the link they follow. The relational model uses a collection of money-ticket style tables, each of which is used for different types of entities. In the 1980s, computing hardware became so powerful that it enabled widespread deployment of relational systems (DBMS + applications). However, in the early 1990s, all large data processing applications were dominated by relational systems, and the situation is still the same by 2015. : IBM DB2, Oracle, MySQL, Microsoft SQL Server is the top-level DBMS [3] The dominant database language, the standardized SQL for the relational model, soon influenced the database languages of other data models.
Object-oriented databases were developed in the 1980s to overcome the inconvenience of object-oriented impedance mismatches, resulting in the term "post-relational" and the development of a hybrid object-relational database.
In the late 2000s, the next generation of relational databases became known as NoSQL databases, introducing high-speed key-value stores and document-oriented databases.
It is a set of information that is integrated and managed for the purpose of sharing and using by many people. It is a collection of one or more logically related data that is highly structured to facilitate the search and update. In other words, it is a collection of data that organizes several data files, eliminates duplication of data items, and structures and memorizes data.
As common data, each user can use the same data differently depending on their application purpose.
Database Features
Real-time accessibility
Continuous change
Simultaneous sharing
Reference to content
Data logical independence
Advantages and disadvantages of databases
Database Benefits
Minimize Data Duplication
Data Sharing
Consistency, Integrity, and Security
Keep your data up-to-date
Data can be standardized
Logical and physical independence of data
Easy data access
Save data storage space
Database Disadvantages
Database Expert Needed
A lot of costs
Difficult to back up and recover data
System complexity
Overload occurs when access is concentrated on a large disk
Database model
The body of this section is the database model.
Some current conceptual logical data models for real database implementations are as follows:
Relational Data Model
The body of this section is the relational model.
The relational data model is the simplest model of the data model. The code I worked for at IBM Labs (E.F.Codd) was proposed in 1970. This model is based on relative mathematical theory. Because the code was a mathematician, I used the concepts of mathematics, especially set theory and logic. In order to develop a data model, a theoretical modeling process, which is described as a table relation, occurs. This is called an entity-relational model. The design process for this relational database is theoretically a real implementation based on relational mathematics. The real world is represented by object relation diagrams, and objects and their relations are drawn with rectangles and lines.
SQL
Structured Query Language (SEQUEL) (Structured English Query Language) was created by IBM Research in 1974 to support an object-relational database, which is based on mathematical relational algebra and relational calculus. I have left. The data model must have a set of operations to manipulate the data. Because it defines the database structure and constraints. In other words, a set of operations of a relational data model is represented by relational algebra, which allows the user to perform multiple queries.
Choosing a database management system
After database design, you should use database management system. There is various database management system options (DBMS).
There is now an 'IMS' database management system that is not widely used but runs under an IBM mainframe environment. IMS is an abbreviation of Information Management System and performs DBMS and DC functions. The DB types managed by IMS are 'DEDB' and 'MSDB'. DEDB is a traditional hierarchical DB with Data Entry DB. MSDB is the main memory DB. DB is resident in main memory when the system is running. It is used for tasks requiring quick access. However, since the access speed is very fast and resides in memory, there are many limitations in capacity and structure.
Select DBMS language
The database language is as follows.
Data definition language (DDL) - Commands such as Create, Alter, and Drop
Data manipulation language (DML) - Select, Insert, Delete, Update ...
Data control language (DCL) - Grant, Revoke, Commit, Rollback ...
transaction
The body of this part is the transaction.
A transaction is a collection of database operations that make up a single logical unit. In order to execute several transactions at the same time, consistency of the database must be ensured. There are modules for concurrency control and recovery control, which are collectively referred to as a transaction management module.
Concurrency control module: Controls the interaction between concurrent transactions to keep the database consistent.
Recovery control module: In order to keep the database consistent, the existing state of the database is maintained even during system failure.
Transaction scheduling has the following three concepts.
Serial scheduling: how to execute transaction operations sequentially for each transaction
Nonserial scheduling: parallel execution of transaction operations with interleaving
Serializable scheduling: When the non-serial scheduling S always has the same result for the serial scheduling SS, say "Sis serializable".
There are protocols to ensure serializable transactions, such as locking methods and timestamps.
Database data structure
Indexing
A database often must meet the following ACID rules:
Atomicity: All operations in a transaction must be performed, or none at all. If the transaction did not execute properly, roll back.
Consistency: All transactions must satisfy the integrity conditions defined in the database.
Isolation: Two transactions cannot affect each other. The value while the transaction is running must not be accessible by other transactions.
Durability: Once a transaction has been successfully completed, the result should remain in the database (even if a system failure occurs).
Concurrency control is a technology that processes transactions securely and satisfies ACID rules.
Demand for the database
In 2012, Korea's domestic DB industry is showing a year-on-year growth in all areas including DB construction market, DB consulting and solution market, and DB service market. In particular, the growth trend will continue in the future as demand for ICT convergence, smart environment expansion, and big data increases in all industries. According to the report, the main factors for the growth of the DB industry are the increase of new demand for analysis and utilization of big data, the increase of DB investment due to the recognition of DB asset value, and the spread of smart-based mobile service.
The relational model originally proposed by Edgar F. Cudd in 1970 began with this tradition, arguing that applications should retrieve data based on their content, not the link they follow. The relational model uses a collection of money-ticket style tables, each of which is used for different types of entities. In the 1980s, computing hardware became so powerful that it enabled widespread deployment of relational systems (DBMS + applications). However, in the early 1990s, all large data processing applications were dominated by relational systems, and the situation is still the same by 2015. : IBM DB2, Oracle, MySQL, Microsoft SQL Server is the top-level DBMS [3] The dominant database language, the standardized SQL for the relational model, soon influenced the database languages of other data models.
Object-oriented databases were developed in the 1980s to overcome the inconvenience of object-oriented impedance mismatches, resulting in the term "post-relational" and the development of a hybrid object-relational database.
In the late 2000s, the next generation of relational databases became known as NoSQL databases, introducing high-speed key-value stores and document-oriented databases.
It is a set of information that is integrated and managed for the purpose of sharing and using by many people. It is a collection of one or more logically related data that is highly structured to facilitate the search and update. In other words, it is a collection of data that organizes several data files, eliminates duplication of data items, and structures and memorizes data.
As common data, each user can use the same data differently depending on their application purpose.
Database Features
Real-time accessibility
Continuous change
Simultaneous sharing
Reference to content
Data logical independence
Advantages and disadvantages of databases
Database Benefits
Minimize Data Duplication
Data Sharing
Consistency, Integrity, and Security
Keep your data up-to-date
Data can be standardized
Logical and physical independence of data
Easy data access
Save data storage space
Database Disadvantages
Database Expert Needed
A lot of costs
Difficult to back up and recover data
System complexity
Overload occurs when access is concentrated on a large disk
Database model
The body of this section is the database model.
Some current conceptual logical data models for real database implementations are as follows:
Relational Data Model
The body of this section is the relational model.
The relational data model is the simplest model of the data model. The code I worked for at IBM Labs (E.F.Codd) was proposed in 1970. This model is based on relative mathematical theory. Because the code was a mathematician, I used the concepts of mathematics, especially set theory and logic. In order to develop a data model, a theoretical modeling process, which is described as a table relation, occurs. This is called an entity-relational model. The design process for this relational database is theoretically a real implementation based on relational mathematics. The real world is represented by object relation diagrams, and objects and their relations are drawn with rectangles and lines.
SQL
Structured Query Language (SEQUEL) (Structured English Query Language) was created by IBM Research in 1974 to support an object-relational database, which is based on mathematical relational algebra and relational calculus. I have left. The data model must have a set of operations to manipulate the data. Because it defines the database structure and constraints. In other words, a set of operations of a relational data model is represented by relational algebra, which allows the user to perform multiple queries.
Choosing a database management system
After database design, you should use database management system. There is various database management system options (DBMS).
There is now an 'IMS' database management system that is not widely used but runs under an IBM mainframe environment. IMS is an abbreviation of Information Management System and performs DBMS and DC functions. The DB types managed by IMS are 'DEDB' and 'MSDB'. DEDB is a traditional hierarchical DB with Data Entry DB. MSDB is the main memory DB. DB is resident in main memory when the system is running. It is used for tasks requiring quick access. However, since the access speed is very fast and resides in memory, there are many limitations in capacity and structure.
Select DBMS language
The database language is as follows.
Data definition language (DDL) - Commands such as Create, Alter, and Drop
Data manipulation language (DML) - Select, Insert, Delete, Update ...
Data control language (DCL) - Grant, Revoke, Commit, Rollback ...
transaction
The body of this part is the transaction.
A transaction is a collection of database operations that make up a single logical unit. In order to execute several transactions at the same time, consistency of the database must be ensured. There are modules for concurrency control and recovery control, which are collectively referred to as a transaction management module.
Concurrency control module: Controls the interaction between concurrent transactions to keep the database consistent.
Recovery control module: In order to keep the database consistent, the existing state of the database is maintained even during system failure.
Transaction scheduling has the following three concepts.
Serial scheduling: how to execute transaction operations sequentially for each transaction
Nonserial scheduling: parallel execution of transaction operations with interleaving
Serializable scheduling: When the non-serial scheduling S always has the same result for the serial scheduling SS, say "Sis serializable".
There are protocols to ensure serializable transactions, such as locking methods and timestamps.
Database data structure
Indexing
A database often must meet the following ACID rules:
Atomicity: All operations in a transaction must be performed, or none at all. If the transaction did not execute properly, roll back.
Consistency: All transactions must satisfy the integrity conditions defined in the database.
Isolation: Two transactions cannot affect each other. The value while the transaction is running must not be accessible by other transactions.
Durability: Once a transaction has been successfully completed, the result should remain in the database (even if a system failure occurs).
Concurrency control is a technology that processes transactions securely and satisfies ACID rules.
Demand for the database
In 2012, Korea's domestic DB industry is showing a year-on-year growth in all areas including DB construction market, DB consulting and solution market, and DB service market. In particular, the growth trend will continue in the future as demand for ICT convergence, smart environment expansion, and big data increases in all industries. According to the report, the main factors for the growth of the DB industry are the increase of new demand for analysis and utilization of big data, the increase of DB investment due to the recognition of DB asset value, and the spread of smart-based mobile service.
No comments:
Post a Comment