国产av日韩一区二区三区精品,成人性爱视频在线观看,国产,欧美,日韩,一区,www.成色av久久成人,2222eeee成人天堂

Table of Contents
MVCC: Make reading and writing not block each other
Gap lock: prevent phantom reading
How does RR achieve "repeatable" and "unillusory reading"?
Home Database Mysql Tutorial How does InnoDB implement Repeatable Read isolation level?

How does InnoDB implement Repeatable Read isolation level?

Jun 14, 2025 am 12:33 AM

InnoDB implements repeatable reads through MVCC and gap lock. MVCC realizes consistent reading through snapshots, and the transaction query results remain unchanged after multiple transactions; gap lock prevents other transactions from inserting data and avoids phantom reading. For example, transaction A first query gets a value of 100, transaction B is modified to 200 and submitted, A is still 100 in query again; and when performing scope query, gap lock prevents other transactions from inserting records. In addition, non-unique index scans may add gap locks by default, and primary key or unique index equivalent queries may not be added, and gap locks can be cancelled by reducing isolation levels or explicit lock control.

How does InnoDB implement Repeatable Read isolation level?

InnoDB implements the Repeatable Read isolation level, which is mainly accomplished through multi-version concurrency control (MVCC) and gap lock . This mechanism can effectively avoid many concurrency problems while ensuring data consistency.


MVCC: Make reading and writing not block each other

InnoDB uses the MVCC mechanism to implement non-locked consistent reads, that is, at the Repeatable Read isolation level, the data seen by the same transaction performing SELECT queries multiple times is consistent, even if other transactions have submitted modifications.

  • Each row of records contains two hidden fields: DB_TRX_ID (the transaction ID of the row was last modified) and DB_ROLL_PTR (points to the undo log in the rollback segment).
  • When a transaction starts, InnoDB creates a consistent read view of transaction visibility, and all subsequent queries use this snapshot to determine which versions of data are visible to the current transaction.
  • So even if other transactions update and data are submitted, these changes will not be seen in the current transaction, thus achieving "repeatable read".

For example:

Suppose that the first time a record is queryed after transaction A starts is 100, then transaction B modifies the record to 200 and submits it. If transaction A querys this record again, it still sees 100 because its snapshot has not changed.


Gap lock: prevent phantom reading

Although MVCC solves the problem of "not repeatable reading", there is another phenomenon called "illusion reading" - that is, the number of result sets changes when queries are queried twice in a transaction. InnoDB uses Gap Lock under Repeatable Read to solve this problem.

  • The gap lock acts on the "gap" between the indexes, preventing other transactions from inserting new records into these gaps.
  • For example, if you execute a range query (such as SELECT * FROM table WHERE id BETWEEN 10 AND 20 FOR UPDATE ), InnoDB will not only add record locks to existing records, but also add gap locks to the intervals from id=10 to id=20.
  • Other transactions cannot insert new records in this interval, thus avoiding phantom reading.

A few explanations:

  • If you are using a non-unique index scan, InnoDB is likely to have a gap lock by default.
  • For primary key or unique index equivalent queries, the gap lock may not take effect.
  • If you do not want to add a gap lock, you can lower the isolation level to Read Committed, or use SELECT ... FOR SHARE or SELECT ... FOR UPDATE to control the lock behavior.

How does RR achieve "repeatable" and "unillusory reading"?

The reason why InnoDB's Repeatable Read isolation level can avoid both non-repeatable reading and fantasy reading is that it combines two mechanisms:

  • MVCC provides a consistent view to ensure that the result of reading the same record multiple times within the same transaction remains unchanged;
  • The gap lock prevents other transactions from inserting new records , preventing phantom readings.

The advantage of this design is that its performance is relatively good and does not require full serialization operations like Serializable.


Basically that's it. The RR-level implementation of InnoDB is actually quite smart. It uses MVCC to handle read consistency and uses gap locks to handle write conflicts. The two work well together. However, if you encounter deadlocks, lock waiting and other problems in actual development, you still need to analyze the usage of SQL and indexes in detail.

The above is the detailed content of How does InnoDB implement Repeatable Read isolation level?. For more information, please follow other related articles on the PHP Chinese website!

Statement of this Website
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn

Hot AI Tools

Undress AI Tool

Undress AI Tool

Undress images for free

Undresser.AI Undress

Undresser.AI Undress

AI-powered app for creating realistic nude photos

AI Clothes Remover

AI Clothes Remover

Online AI tool for removing clothes from photos.

Clothoff.io

Clothoff.io

AI clothes remover

Video Face Swap

Video Face Swap

Swap faces in any video effortlessly with our completely free AI face swap tool!

Hot Tools

Notepad++7.3.1

Notepad++7.3.1

Easy-to-use and free code editor

SublimeText3 Chinese version

SublimeText3 Chinese version

Chinese version, very easy to use

Zend Studio 13.0.1

Zend Studio 13.0.1

Powerful PHP integrated development environment

Dreamweaver CS6

Dreamweaver CS6

Visual web development tools

SublimeText3 Mac version

SublimeText3 Mac version

God-level code editing software (SublimeText3)

What is the default username and password for MySQL? What is the default username and password for MySQL? Jun 13, 2025 am 12:34 AM

The default user name of MySQL is usually 'root', but the password varies according to the installation environment; in some Linux distributions, the root account may be authenticated by auth_socket plug-in and cannot log in with the password; when installing tools such as XAMPP or WAMP under Windows, root users usually have no password or use common passwords such as root, mysql, etc.; if you forget the password, you can reset it by stopping the MySQL service, starting in --skip-grant-tables mode, updating the mysql.user table to set a new password and restarting the service; note that the MySQL8.0 version requires additional authentication plug-ins.

What is GTID (Global Transaction Identifier) and what are its advantages? What is GTID (Global Transaction Identifier) and what are its advantages? Jun 19, 2025 am 01:03 AM

GTID (Global Transaction Identifier) ??solves the complexity of replication and failover in MySQL databases by assigning a unique identity to each transaction. 1. It simplifies replication management, automatically handles log files and locations, allowing slave servers to request transactions based on the last executed GTID. 2. Ensure consistency across servers, ensure that each transaction is applied only once on each server, and avoid data inconsistency. 3. Improve troubleshooting efficiency. GTID includes server UUID and serial number, which is convenient for tracking transaction flow and accurately locate problems. These three core advantages make MySQL replication more robust and easy to manage, significantly improving system reliability and data integrity.

How to change or reset the MySQL root user password? How to change or reset the MySQL root user password? Jun 13, 2025 am 12:33 AM

There are three ways to modify or reset MySQLroot user password: 1. Use the ALTERUSER command to modify existing passwords, and execute the corresponding statement after logging in; 2. If you forget your password, you need to stop the service and start it in --skip-grant-tables mode before modifying; 3. The mysqladmin command can be used to modify it directly by modifying it. Each method is suitable for different scenarios and the operation sequence must not be messed up. After the modification is completed, verification must be made and permission protection must be paid attention to.

What is a typical process for MySQL master failover? What is a typical process for MySQL master failover? Jun 19, 2025 am 01:06 AM

MySQL main library failover mainly includes four steps. 1. Fault detection: Regularly check the main library process, connection status and simple query to determine whether it is downtime, set up a retry mechanism to avoid misjudgment, and can use tools such as MHA, Orchestrator or Keepalived to assist in detection; 2. Select the new main library: select the most suitable slave library to replace it according to the data synchronization progress (Seconds_Behind_Master), binlog data integrity, network delay and load conditions, and perform data compensation or manual intervention if necessary; 3. Switch topology: Point other slave libraries to the new master library, execute RESETMASTER or enable GTID, update the VIP, DNS or proxy configuration to

How to connect to a MySQL database using the command line? How to connect to a MySQL database using the command line? Jun 19, 2025 am 01:05 AM

The steps to connect to the MySQL database are as follows: 1. Use the basic command format mysql-u username-p-h host address to connect, enter the username and password to log in; 2. If you need to directly enter the specified database, you can add the database name after the command, such as mysql-uroot-pmyproject; 3. If the port is not the default 3306, you need to add the -P parameter to specify the port number, such as mysql-uroot-p-h192.168.1.100-P3307; In addition, if you encounter a password error, you can re-enter it. If the connection fails, check the network, firewall or permission settings. If the client is missing, you can install mysql-client on Linux through the package manager. Master these commands

How does InnoDB implement Repeatable Read isolation level? How does InnoDB implement Repeatable Read isolation level? Jun 14, 2025 am 12:33 AM

InnoDB implements repeatable reads through MVCC and gap lock. MVCC realizes consistent reading through snapshots, and the transaction query results remain unchanged after multiple transactions; gap lock prevents other transactions from inserting data and avoids phantom reading. For example, transaction A first query gets a value of 100, transaction B is modified to 200 and submitted, A is still 100 in query again; and when performing scope query, gap lock prevents other transactions from inserting records. In addition, non-unique index scans may add gap locks by default, and primary key or unique index equivalent queries may not be added, and gap locks can be cancelled by reducing isolation levels or explicit lock control.

How to alter a large table without locking it (Online DDL)? How to alter a large table without locking it (Online DDL)? Jun 14, 2025 am 12:36 AM

Toalteralargeproductiontablewithoutlonglocks,useonlineDDLtechniques.1)IdentifyifyourALTERoperationisfast(e.g.,adding/droppingcolumns,modifyingNULL/NOTNULL)orslow(e.g.,changingdatatypes,reorderingcolumns,addingindexesonlargedata).2)Usedatabase-specifi

What are the transaction isolation levels in MySQL, and which is the default? What are the transaction isolation levels in MySQL, and which is the default? Jun 23, 2025 pm 03:05 PM

MySQL's default transaction isolation level is RepeatableRead, which prevents dirty reads and non-repeatable reads through MVCC and gap locks, and avoids phantom reading in most cases; other major levels include read uncommitted (ReadUncommitted), allowing dirty reads but the fastest performance, 1. Read Committed (ReadCommitted) ensures that the submitted data is read but may encounter non-repeatable reads and phantom readings, 2. RepeatableRead default level ensures that multiple reads within the transaction are consistent, 3. Serialization (Serializable) the highest level, prevents other transactions from modifying data through locks, ensuring data integrity but sacrificing performance;

See all articles