[learning notes] classification of locks in the database

From the sharing of locks / From the perspective of database management , Locks can be divided into shared locks and exclusive locks .

Shared lock （S lock ）： Also called read lock , For read-only operations that do not modify data , Such as select sentence . When a transaction applies a shared lock to data , Other transactions can only add shared locks to the same transactions , Instead of adding an exclusive lock . Transactions that acquire shared locks can only read data , You can't modify the data .

Exclusive lock （X lock ）： Also known as write lock , Operations for modifying data , Such as insert、update、delete sentence . When a transaction adds an exclusive lock to the data , Other transactions can no longer lock the data . Transactions that acquire exclusive locks can read data , It can also modify data .

From the locking mode / Programmer perspective , Locks can be divided into optimistic locks and pessimistic locks .

Optimism lock ： Take an optimistic attitude when locking , That is, there will be no conflict , Do it first , If there is no conflict, you will succeed directly , If the conflict is resolved .

Pessimistic locking ： Adopt a pessimistic attitude when locking , That is, there must be conflict , No matter there is conflict , Lock them first , Then operate .

From the granularity of shackles , Locks can be divided into row locks and table locks .

Row lock ： Lock a row of the database table , Spending big , Small lock size , The probability of conflict is small , High concurrency .

Table locks ： Lock the whole table of the database , Low overhead , Big lock size , There is a high probability of conflict , Low concurrency .