Chapter 10
Transaction Management and
Concurrency Control
Learning Objectives
• After completing this chapter, you will be able to:
• Describe the database transaction management process
• Identify the four properties of a database transaction
• Explain concurrency control and its role in maintaining database integrity
• Describe how locking methods are used for concurrency control
• Describe how stamping methods are used for concurrency control
• Describe how optimistic methods are used for concurrency control
• List and explain the ANSI levels of transaction isolation
• Describe the role of database recovery management in maintaining database integrity
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 2
for classroom use.
What is a Transaction?
• A logical unit of work that must be entirely completed or aborted
• Consists of:
- SELECT statement
- Series of related UPDATE statements
- Series of INSERT statements
- Combination of SELECT, UPDATE, and INSERT statements
• Consistent database state
• All data integrity constraints are satisfied
• Must begin with the database in a known consistent state to ensure consistency
• Most are formed by two or more database requests
- Database requests: equivalent of a single SQL statement in an application program or
transaction
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 3
for classroom use.
Evaluating Transaction Results
• Not all transactions update the database
• SQL code represents a transaction because it accesses the database
• Improper or incomplete transactions can have devastating effect on database
integrity
• Users can define enforceable constraints based on business rules
• Other integrity rules are automatically enforced by the DBMS
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 4
for classroom use.
Transaction Properties
• Atomicity
• All operations of a transaction must be completed; if not the transaction is aborted
• Consistency
• Permanence of database’s consistent state
• Isolation
• Data used during transaction cannot be used by second transaction until the first is
completed
• Durability
• Ensures that once transactions are committed they cannot be undone or lost
• Serializability
• Ensures that the schedule for the concurrent execution of several transactions should
yield consistent results
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 5
for classroom use.
Transaction Management with SQL
• SQL statements that provide transaction support:
• COMMIT
• ROLLBACK
• Transaction sequence must continue until one of four events occur:
• COMMIT statement is reached
• ROLLBACK statement is reached
• End of program is reached
• Program is abnormally terminated
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 6
for classroom use.
The Transaction Log
• Keeps track of all transactions that update the database
• DBMS uses the information stored in a log for (a):
- Recovery requirement triggered by a ROLLBACK statement
- Program’s abnormal termination
- System failure
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 7
for classroom use.
Concurrency Control
• Coordination of the simultaneous transactions execution in a multiuser
database system
• Objective: ensures serializability of transactions in a multiuser database environment
• Important because the simultaneous execution of transactions over a shared
database can create several data integrity and consistency problems
• Three main problems are lost updates, uncommitted data, and inconsistent retrievals
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 8
for classroom use.
Problems in Concurrency Control
• Lost update
• Occurs in two concurrent transactions when:
- Same data element is updated
- One of the updates is lost
• Uncommitted data
• Occurs when:
- Two transactions are executed concurrently
- First transaction is rolled back after the second transaction has already accessed uncommitted
data
• Inconsistent retrievals
• Occurs when:
- A transaction accesses data before and after one or more other transactions finish working
with such data
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 9
for classroom use.
The Scheduler
• Establishes the order in which the operations are executed within concurrent
transactions
• Interleaves the execution of database operations to ensure serializability and isolation
of transactions
• Bases actions on concurrent control algorithms
• Determines appropriate order
• Creates serialization schedule
• Serializable schedule: interleaved execution of transactions yields the same results as
the serial execution of the transactions
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 10
for classroom use.
Concurrency Control with Locking Methods
• Locking methods facilitate isolation of data items used in concurrently
executing transactions
• Lock: guarantees exclusive use of a data item to a current transaction
• Pessimistic locking: use of locks based on the assumption that conflict between
transactions is likely
• Lock manager: responsible for assigning and policing the locks used by the
transactions
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 11
for classroom use.
Lock Granularity (1 of 5)
• Indicates the level of lock use
• Database-level lock
• Table-level lock
• Page-level lock
- Page or diskpage: directly addressable section of a disk
• Row-level lock
• Field-level lock
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 12
for classroom use.
Lock Granularity (2 of 5)
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 13
for classroom use.
Lock Granularity (3 of 5)
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 14
for classroom use.
Lock Granularity (4 of 5)
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 15
for classroom use.
Lock Granularity (5 of 5)
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 16
for classroom use.
Lock Types (1 of 2)
• Binary lock
• Two states: locked (1) and unlocked (0)
- If an object is locked by a transaction, no other transaction can use that object
- If an object is unlocked, any transaction can lock the object for its use
• Exclusive lock
• Access is reserved for the transaction that locked the object
• Shared lock
• Concurrent transactions are granted read access on the basis of a common lock
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 17
for classroom use.
Lock Types (2 of 2)
• Problems using locks
• Resulting transaction schedule might not be serializable
• Schedule might create deadlocks
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 18
for classroom use.
Two-Phase Locking to Ensure Serializability (1 of 3)
• Defines how transactions acquire and relinquish locks
• Guarantees serializability but does not prevent deadlocks
• Phases
• Growing phase: transaction acquires all required locks without unlocking any data
• Shrinking phase: transaction releases all locks and cannot obtain any new lock
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 19
for classroom use.
Two-Phase Locking to Ensure Serializability (2 of 3)
• Governing rules
• Two transactions cannot have conflicting locks
• No unlock operation can precede a lock operation in the same transaction
• No data are affected until all locks are obtained
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 20
for classroom use.
Two-Phase Locking to Ensure Serializability (3 of 3)
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 21
for classroom use.
Deadlocks
• Occur when two transactions wait indefinitely for each other to unlock data
• Also known as deadly embrace
• Control techniques
• Deadlock prevention
• Deadlock detection
• Deadlock avoidance
• Choice of deadlock control method depends on database environment
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 22
for classroom use.
Concurrency Control with Time Stamping Methods (1 of 2)
• Time stamping assigns global, unique time stamp to each transaction
• Produces explicit order in which transactions are submitted to DBMS
• Properties
• Uniqueness: ensures no equal time stamp values exist
• Monotonicity: ensures time stamp values always increases
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 23
for classroom use.
Concurrency Control with Time Stamping Methods (2 of 2)
• Disadvantages
• Each value stored in the database requires two additional stamp fields
• Increases memory needs
• Increases the database’s processing overhead
• Demands a lot of system resources
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 24
for classroom use.
Wait/Die and Wound/Wait Schemes (1 of 2)
• Wait/die
• A concurrency control scheme in which an older transaction must wait for the
younger transaction to complete and release the locks before requesting the locks
itself
- Otherwise, the newer transaction dies and is rescheduled
• Wound/wait
• A concurrency control scheme in which an older transaction can request the lock,
preempt the younger transaction, and reschedule it
- Otherwise, the newer transaction waits until the older transaction finishes
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 25
for classroom use.
Wait/Die and Wound/Wait Schemes (2 of 2)
Wait/Die and
Wound/Wait
Concurrency Control
Schemes
Transaction Transaction Wait/Die Scheme Wound/Wait Scheme
Requesting Lock Owning Lock
T1 (11548789) T2 (19562545) • T1 waits until T2 is • T1 preempts (rolls back)
completed and T2 T2.
releases its locks. • T2 is rescheduled using
the same time stamp.
T2 (19562545) T1 (11548789) • T2 dies (rolls back). • T2 waits until T1 is
• T2 is rescheduled completed and T1
using the same time releases its locks.
stamp.
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 26
for classroom use.
Concurrency Control with Optimistic Methods (1 of 2)
• Optimistic approach: based on the assumption that the majority of database
operations do not conflict
• Does not require locking or time stamping techniques
• Transaction is executed without restrictions until it is committed
• Phases of optimistic approach
• Read
• Validation
• Write
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 27
for classroom use.
Concurrency Control with Optimistic Methods (2 of 2)
• Read phase
• Transaction:
- Reads the database
- Executes the needed computations
- Makes the updates to a private copy of the database values
• Validation phase
• Transaction is validated to ensure that the changes made will not affect the integrity
and consistency of the database
• Write phase
• Changes are permanently applied to the database
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 28
for classroom use.
ANSI Levels of Transaction Isolation (1 of 2)
• The ANSI SQL standard (1992) defines transaction management based on
transaction isolation levels
• Transaction isolation levels refer to the degree to which transaction data is “protected
or isolated” from other concurrent transactions
• Transaction isolation levels are described by the type of “reads” that a
transaction allows or not
• Dirty read: transaction can read data that is not yet committed
• Nonrepeatable read: transaction reads a given row at time t1, and then it reads the
same row at time t2, yielding different results
- The original row may have been updated or deleted
• Phantom read: transaction executes a query at time t1, and then it runs the same
query at time t2, yielding additional rows that satisfy the query
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 29
for classroom use.
ANSI Levels of Transaction Isolation (2 of 2)
• Read Uncommitted will read uncommitted data from other transactions
• Increases transaction performance but at the cost of data consistency
• Read Committed forces transactions to read only committed data
• Default mode of operation for most databases
• Repeatable Read isolation level ensures that queries return consistent results
• Uses shared locks to ensure other transactions do not update a row after the original
query reads it
• Serializable isolation level is the most restrictive level defined by the ANSI SQL
standard
• Deadlocks are still always possible
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 30
for classroom use.
Database Recovery Management
• Database recovery: restores database from a given state to a previously
consistent state
• Recovery transactions are based on the atomic transaction property
• All portions of a transaction must be treated as a single logical unit of work
- If transaction operation cannot be completed:
• Transaction must be aborted
• Changes to database must be rolled back
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 31
for classroom use.
Transaction Recovery (1 of 3)
• Concepts that affect the recovery process
• Write-ahead log protocol
- Ensures that transaction logs are always written before the data are updated
• Redundant transaction logs
- Ensure that a physical disk failure will not impair the DBMS’s ability to recover data
• Buffers
- Temporary storage areas in a primary memory used to speed up disk operations
• Checkpoints
- Allows DBMS to write all its updated buffers in memory to disk
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 32
for classroom use.
Transaction Recovery (2 of 3)
• Techniques used in transaction recovery procedures
• Deferred-write technique or deferred update
- Transaction operations do not immediately update the physical database
- Only transaction log is updated
• Write-through technique or immediate update
- Database is immediately updated by transaction operations during transaction’s execution
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 33
for classroom use.
Transaction Recovery (3 of 3)
• Recovery process steps
• Identify the last check point in the transaction log
- If transaction was committed before the last check point nothing needs to be done
- If transaction was committed after the last check point the transaction log is used to redo the
transaction
- If transaction had a ROLLBACK operation after the last check point the DBMS uses the
transaction log records to ROLLBACK or undo the operations, using the “before” values in the
transaction log
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 34
for classroom use.
Summary (1 of 2)
• A transaction is a sequence of database operations that access the database
• Transactions have four main properties: atomicity, consistency, isolation, and
durability
• SQL provides support for transactions through the use of two statements:
COMMIT, which saves changes to disk, and ROLLBACK, which restores the
previous database state
• Concurrency control coordinates the simultaneous execution of transactions
• A lock guarantees unique access to a data item by a transaction
• Serializability of schedules is guaranteed through the use of two-phase locking
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 35
for classroom use.
Summary (2 of 2)
• Concurrency control with time stamping methods assigns a unique time stamp
to each transaction and schedules the execution of conflicting transactions in
time stamp order
• Concurrency control with optimistic methods assumes that the majority of
database transactions do not conflict and that transactions are executed
concurrently, using private, temporary copies of the data
• Database recovery restores the database from a given state to a previous
consistent state
© 2019 Cengage. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-protected website 36
for classroom use.
