8/11/2019 Lec 1 Transaction Processing Concepts
1/87
Database Systems
Introduction to TransactionProcessing Concepts and
TheoryDr. Ejaz Ahmed
1
Read comments and additional details
given with some slides
8/11/2019 Lec 1 Transaction Processing Concepts
2/87
Outline
Introduction to transaction processing Transaction and system concepts
Desirable properties of transactions
Schedules and recoverability Schedules and Serializability
Transaction support in SQL
Summary
2
8/11/2019 Lec 1 Transaction Processing Concepts
3/87
Introduction to Transaction
Processing
Single-user VS multi-user systems A DBMS is single-user if at most one user
can use the system at a time
A DBMS is multi-user if many users can use
the system concurrently
Problem
How to make the simultaneous interactions
of multiple users with the database safe,consistent, correct, and efficient?
3
8/11/2019 Lec 1 Transaction Processing Concepts
4/87
Introduction to Transaction
Processing Computing systems
Single-processor computer system
Multiprogramming
Inter-leaved Execution
Pseudo-parallel processing
Multi-processor computer system
Parallel processing
4
8/11/2019 Lec 1 Transaction Processing Concepts
5/87
Concurrent Transactions
5
Interleaved processing
(Single processor)Parallel processing
(Two or more processors)
t1 t2 t1 t2
time
CPU1
CPU2A
B
A
B
CPU1A
B
8/11/2019 Lec 1 Transaction Processing Concepts
6/87
What is a Transaction? A transaction T is a logical unit of database processing
that includes one or more database access operations Embedded within an application program
Specified interactively (e.g., via SQL)
Transaction boundaries:
Begin/end transaction Types of transactions
Read transaction
write transaction
Read-set of T: all data items that transaction T reads
Write-set of T: all data items that transaction T writes
6
8/11/2019 Lec 1 Transaction Processing Concepts
7/87
Role of Transactions
7
Transaction processing (TP) systems underlie the online
applications. a transaction rate of 15 simple Debit Credit
transactions per second was common, 50 TPS was a goodsystem, 100 TPS was fast, and a 400 TPS system was
characterized as being a very lean and mean system [24].
Teradata capabilities, usage of parallel databases,configuration and tuning etc.
8/11/2019 Lec 1 Transaction Processing Concepts
8/87
A Transaction: An Informal Example
Transfer 400,000 from checking account tosavings account
For a user it is one activity
To database:
Read balance of checking account: read( X)
Read balance of savings account: read (Y)
Subtract 400,000 from X
Add 400,000 to Y Write new value of X back to disk
Write new value of Y back to disk
8
8/11/2019 Lec 1 Transaction Processing Concepts
9/87
Transactions Example
Usage of ATM to draw money, to collect bills
with dispenser
9
8/11/2019 Lec 1 Transaction Processing Concepts
10/87
Database Read and Write
Operations A database is represented as a collection of named
data items Read-item (X)
1. Find the address of the disk block that contains itemX
2. Copy the disk block into a buffer in main memory
3. Copy the item X from the buffer to the program variable namedX
Write-item (X)1. Find the address of the disk block that contains itemX.
2. Copy that disk block into a buffer in main memory
3. Copy itemX from the program variable namedX into its correctlocation in the buffer.
4. Store the updated block from the buffer back to disk (eitherimmediately or at some later point in time).
10
8/11/2019 Lec 1 Transaction Processing Concepts
11/87
A Transaction: A Formal
ExampleT1
read_item(X);
read_item(Y);X:=X - 400000;
Y:=Y + 400000;
write _item(X);
write_item(Y);
11
t0
tk
8/11/2019 Lec 1 Transaction Processing Concepts
12/87
Introduction to Transaction
Processing(Cont.)
Why concurrency control is needed? Three problems are
1. The lost update problem
2. The temporary update (dirty read) problem3. Incorrect summary problem
12
8/11/2019 Lec 1 Transaction Processing Concepts
13/87
Operations & Consistent States
Database Operation (read and write)
Non-database Operation (calculations/
formula)
Database Consistent State
Database Inconsistent State
Chapter 17-13
8/11/2019 Lec 1 Transaction Processing Concepts
14/87
9-14
Concurrent Transaction
Concurrent transactionsrefer to two or more
transactions that appear to users as they are being
processed against a database at the same time
In reality, CPU can execute only one instruction at a time
Transactions are interleavedmeaning that the operating
system quickly switches CPU services among tasks so that
some portion of each of them is carried out in a given interval
Concurrency problems: lost update and inconsistent
reads
8/11/2019 Lec 1 Transaction Processing Concepts
15/87
9-15
Concurrent Transaction Processing
8/11/2019 Lec 1 Transaction Processing Concepts
16/87
9-16
Lost-Update Problem
8/11/2019 Lec 1 Transaction Processing Concepts
17/87
Transaction Examples
Chapter 17-17
$10 could be lost: inconsistent state
Transaction is committed or is aborted, DB is rollback or undo
8/11/2019 Lec 1 Transaction Processing Concepts
18/87
PL/SQL Transaction Example
18
8/11/2019 Lec 1 Transaction Processing Concepts
19/87
9-19
8/11/2019 Lec 1 Transaction Processing Concepts
20/87
ExampleThe Lost Update
Problem
Chapter 17-20
8/11/2019 Lec 1 Transaction Processing Concepts
21/87
Chapter 17-21
FIGURE 17.2
Two sample transactions. (a) Transaction T1.
(b) Transaction T2.
8/11/2019 Lec 1 Transaction Processing Concepts
22/87
Chapter 17-22
Some problems that occur when concurrent
execution is uncontrolled. (b) The temporary updateproblem (Dirty Read).
8/11/2019 Lec 1 Transaction Processing Concepts
23/87
Lost Update Problem
T1read_item(X);
X:=X - N;
write_item(X);
read_item(Y);
Y:=Y + N;
write_item(Y);
23
T2
read_item(X);
X:=X+M;
write_item(X);
time
8/11/2019 Lec 1 Transaction Processing Concepts
24/87
Temporary Update (Dirty Read)
T1read_item(X);
X:=X - N;
write_item(X);
read_item(Y);
T1 fai ls and aborts24
T2
read_item(X);
X:=X+M;
write_item(X);
time
8/11/2019 Lec 1 Transaction Processing Concepts
25/87
25
FIGURE 17.3 (continued)
Some problems that occur when concurrent execution is
uncontrolled. (c) The incorrect summary problem
(inconsistent).
8/11/2019 Lec 1 Transaction Processing Concepts
26/87
Incorrect Summary ProblemT1
read_item(X);
X:=X-N;
write_item(X);
read_item(Y);
Y=Y+N
Write_item(Y)
26
T2
sum:=0;
read_item(A);
sum:=sum+A;
read_item(X);
sum:=sum+X;
read_item(Y);
sum:=sum+Y
time
Wh t C G W ?
8/11/2019 Lec 1 Transaction Processing Concepts
27/87
What Can Go Wrong? System may crash before data is written back to
disk
= Problem of atomicity
Some transaction is modifying shared data whileanother transaction is ongoing (or vice versa)= Problem of serialization and isolation
System may not be able to obtain one or more ofthe data items
System may not be able to write one or more ofthe data items= Problems of atomicity
DBMS has a Concurrency Controlsubsytem toassure database remains in consistent state
despite concurrent execution of transactions 27
8/11/2019 Lec 1 Transaction Processing Concepts
28/87
Other Problems System failures may occur
Types of failures:
System crash
Transaction or system error
Local errors Concurrency control enforcement
Disk failure
Physical failures
DBMS has a RecoverySubsystem to
protect database against system
failures
28
8/11/2019 Lec 1 Transaction Processing Concepts
29/87
Introduction to Transaction
Processing(Cont.) Why recovery is needed?
1. A computer failure (system crash)
2. A transaction or system error
3. Local errors or exception conditions detected
by the transaction
4. Concurrency control enforcement
5. Disk failure
6. Physical problems and catastrophes
29
8/11/2019 Lec 1 Transaction Processing Concepts
30/87
Chapter 17-30
Why recovery is needed:(What causes a Transaction to fail)
1. A computer failure (system crash):A hardware orsoftware error occurs in the computer system duringtransaction execution. If the hardware crashes, thecontents of the computers internal memory may belost.
2. A transaction or system error :Some operation in thetransaction may cause it to fail, such as integer overflow
or division by zero. Transaction failure may also occurbecause of erroneous parameter values or because ofa logical programming error. In addition, the user mayinterrupt the transaction during its execution.
Introduction to Transaction Processing (11)
8/11/2019 Lec 1 Transaction Processing Concepts
31/87
Chapter 17-31
Why recovery is needed (cont.):3.Local errors or exception conditionsdetected by the
transaction:
- certain conditions necessitate cancellation of the transaction.For example, data for the transaction may not be found. Acondition, such as insufficient account balance in a banking
database, may cause a transaction, such as a fund withdrawalfrom that account, to be canceled.
- a programmed abort in the transaction causes it to fail.
4. Concurrency control enforcement:The concurrency controlmethod may decide to abort the transaction, to be restarted later,
because it violates serializability or because several transactionsare in a state of deadlock (see Chapter 18).
Introduction to Transaction Processing (12)
8/11/2019 Lec 1 Transaction Processing Concepts
32/87
Chapter 17-32
Why recovery is needed (cont.):5. Disk failure:Some disk blocks may lose their data
because of a read or write malfunction or because of
a disk read/write head crash. This may happen during
a read or a write operation of the transaction.6. Physical problems and catastrophes:This refers
to an endless list of problems that includes power or
air-conditioning failure, fire, theft, sabotage,
overwriting disks or tapes by mistake, and mountingof a wrong tape by the operator.
Introduction to Transaction Processing (13)
T ti d S t
8/11/2019 Lec 1 Transaction Processing Concepts
33/87
Transaction and System
Concepts Transaction states
BEGIN_TRANSACTION: marks start of transaction
READor WRITE: two possible operations on the
data
END_TRANSACTION: marks the end of the read orwrite operations; start checking whether
everything went according to plan
COMIT_TRANSACTION: signals successful end of
transaction; changes can be committed to DB Partially committed
ROLLBACK(or ABORT): signals unsuccessful end of
transaction, changes applied to DB must be
undone
33
8/11/2019 Lec 1 Transaction Processing Concepts
34/87
Transaction States: A state transition diagram
34
8/11/2019 Lec 1 Transaction Processing Concepts
35/87
Chapter 17-35
Transaction and System Concepts (3)
Recovery manager keeps track of the following operations
(cont):
commit_transaction:This signals a successful endof
the transaction so that any changes (updates) executed
by the transaction can be safely committedto the
database and will not be undone.
rollback (or abort): This signals that the transaction
has ended unsuccessfully,so that any changes oreffects that the transaction may have applied to the
database must be undone.
8/11/2019 Lec 1 Transaction Processing Concepts
36/87
Chapter 17-36
Transaction and System Concepts (4)
Recovery techniques use the following operators:
undo:Similar to rollback except that it applies to
a single operation rather than to a wholetransaction.
redo:This specifies that certain transaction
operationsmust be redoneto ensure that all the
operations of a committed transaction havebeen applied successfully to the database.
8/11/2019 Lec 1 Transaction Processing Concepts
37/87
Chapter 17-37
Transaction and System Concepts (6)
The System Log Log or Journal: The log keeps track of all transaction
operations that affect the values of database items. This
information may be needed to permit recovery fromtransaction failures. The log is kept on disk, so it is notaffected by any type of failure except for disk orcatastrophic failure. In addition, the log is periodicallybacked up to archival storage (tape) to guard against
such catastrophic failures. T in the following discussion refers to a unique
transaction-idthat is generated automatically by thesystem and is used to identify each transaction:
Th S t L
8/11/2019 Lec 1 Transaction Processing Concepts
38/87
The System Log Transactionid
System log Multiple record-type file
Log is kept on disk
Periodically backed up
Log records1. [start_transaction, T]2. [write_item, T,X,old_value,new_value]:
3. [read_item, T,X]
4. [commit,T]5. [abort,T]
6. [checkpoint]
Commit point of a transaction
38
8/11/2019 Lec 1 Transaction Processing Concepts
39/87
How is the Log File Used? All permanent changes to data are recorded
Possible to undo changes to data After crash, search log backwards until find
last checkpoint
Know that beyond this point, effects of transaction
are permanently recorded
Need to either redoor undoeverything that
happened since last checkpoint
Undo: When transaction only partially completed
(before crash)
Redo: Transaction completed but we are unsure
whether data was written to disk
39
8/11/2019 Lec 1 Transaction Processing Concepts
40/87
Example of a Transaction Log
TransID Table RowID column Before After
------- --------- ------ ------- ------ ------
100 **start**
100 EMPLOYEE 1005472 Salary 8000 8250
100 PROJECT 1423PRJ PrjRate 253 256
100 **end committed**
Chapter 17-40
8/11/2019 Lec 1 Transaction Processing Concepts
41/87
A Sample SQL TransactionEXEC SQLWHENEVER SQLERROR GOTO UNDO;
EXEC SQLSET TRANSACTION
READ WRITE
DIAGONOSTIC SIZE 5
ISOLATION LEVEL SERIALIZABLE;
EXEC SQLINSERT INTO
EMPLOYEE(FNAME, LNAME, SSN, DNO, SALARY)VALUES (Ali, Al-Fares, 991004321, 2, 35000)
EXEC SQLUPDATE EMPLOYEE
SET SALARY = SALARY * 1.1 WHERE DNO = 2;
EXEC SQLCOMMIT;
GOTO END_T;UNDO: EXEC SQLROLLBACK;
END_T: ;
41
8/11/2019 Lec 1 Transaction Processing Concepts
42/87
Checkpoints The point of synchronization between the database and
the transaction log file is called checkpoint.
Database recovery is performed using transaction log.
How far to go back in the transaction log to search in
case of failure.
Redoing transactions which are already Written to
databaseTime consuming & wasteful.
To find a point before which transactions are done
correctly and safelychecking point
In checkpoint technique, all buffers are force written to
secondary storage.
42
8/11/2019 Lec 1 Transaction Processing Concepts
43/87
Checkpoint technique ... Is used to limit
the volume of log information
amount of searching and
subsequent processing that is needed to carry out on the
transaction log file.
Forcing a Checkpoint: Example The following statement forces a
checkpoint:
ALTER SYSTEM CHECKPOINT;
Enabling Resource Limits: Example This ALTER SYSTEM
statement dynamically enables resource limits:
ALTER SYSTEM SET RESOURCE_LIMIT = TRUE;
43
Oracle CKPT
8/11/2019 Lec 1 Transaction Processing Concepts
44/87
Oracle CKPT Oracle Database uses the following types of checkpoints (CKPT)
Consistent database shutdown
ALTER SYSTEM CHECKPOINT statementOnline redo log switch
ALTER DATABASE BEGIN BACKUP statement
A tablespace checkpoint is a set of data file checkpoints, one for
each data file in the tablespace. These checkpoints occur in a
variety of situations, including making a tablespace read-only or
taking it offline normal, shrinking a data file, or executing
ALTER TABLESPACE BEGIN BACKUP.
44
Desirable Properties of
8/11/2019 Lec 1 Transaction Processing Concepts
45/87
Desirable Properties of
Transactions
ACID properties1. AtomicityA transaction is an atomic unit of processing; it is either
performed in its entirety or not performed at all.
2. Consistency preservation
A transaction is consistency preserving if its completeexecution takes the database from one consistent state to
another
3. Isolation
The execution of a transaction should not be interfered with by
any other transactions executing concurrently4. Durability
The changes applied to the database by a committed
transaction must persist in the database. These changes must
not be lost because of any failure
45
D i bl P ti f
8/11/2019 Lec 1 Transaction Processing Concepts
46/87
Desirable Properties of
Transactions Atomicity
Responsibility of transaction processing and recovery
subsystems of the DBMS
Consistency
Preservation of consistency is the responsibility ofprogrammers
Each transaction is assumed to take database from one
consistent state to another consistent state
Isolation
Enforced by the concurrency control subsystem of theDBMS
Durability
Responsibility of the recovery subsystems of the DBMS
46
8/11/2019 Lec 1 Transaction Processing Concepts
47/87
Transaction Processing
We have discussed that
Multiple transactions can be executed
concurrently by interleaving their operations
Schedule
Ordering of execution of operations from
various transactions T1, T2, , Tnis called a
schedule S
47
8/11/2019 Lec 1 Transaction Processing Concepts
48/87
Schedules and Recoverability Definition of Schedule (or history)
Schedule S of n transactions T1,T2, , Tnis an ordering of the
operations of the transactionssubject to the constraint that, foreach transaction Tithatparticipates in S, the operations of
Tiin S must appear in the sameorder in which they occur in Ti.
48
8/11/2019 Lec 1 Transaction Processing Concepts
49/87
Example of a Schedule
Transaction T1: r1(X); w1(X); r1(Y); w1(Y);c1
Transaction T2: r2(X); w2(X); c2
A schedule, S:
r1(X); r2(X); w1(X); r1(Y); w2(X); w1(Y); c1;
c2
49
8/11/2019 Lec 1 Transaction Processing Concepts
50/87
Conflicts Two operations conflictif they satisfy ALL
three conditions:
1. they belong to different transactions AND
2. they access the same item AND
3. at least one is a write_item()operation
Example.:
S: r1(X); r2(X); w1(X); r1(Y); w2(X); w1(Y);
50
conflicts
S h d l f T ti
8/11/2019 Lec 1 Transaction Processing Concepts
51/87
Schedules of Transactions Complete schedule
A schedule S of n transactions T1, T2, ..., Tn, is
said to be a complete schedule if the followingconditions hold:
The operations in S are exactly those operations in T1, T2,..., Tn including a commit or abort operation as the lastoperation for each transaction in the schedule.
For any pair of operations from the same transaction Ti,their order of appearance in Sis the same as their order ofappearance in Ti.
For any two conflicting operations, one of the two mustoccur before the other in the schedule
51
8/11/2019 Lec 1 Transaction Processing Concepts
52/87
Serializability of Schedules
Serial Schedule
Non-serial schedule
Serializable schedule
Conflict-serializable schedule
View-serializable schedule
File Attached: Ch17-18doc
52
8/11/2019 Lec 1 Transaction Processing Concepts
53/87
Chapter 17-53
Characterizing Schedules
based on Serializability (2)
Result equivalent: Two schedules are called
result equivalent if they produce the same final
state of the database.
Conflict equivalent: Two schedules are said to
be conflict equivalent if the order of any two
conflicting operations is the same in both
schedules. Conflict serializable: A schedule S is said to be
conflict serializable if it is conflict equivalent to
some serial schedule S.
8/11/2019 Lec 1 Transaction Processing Concepts
54/87
54
8/11/2019 Lec 1 Transaction Processing Concepts
55/87
Chapter 17-55
Characterizing Schedules
based on Serializability (3)
Being serializable is not the same as being
serial
Being serializable implies that the schedule is a
correct schedule.
It will leave the database in a consistent state.
The interleaving is appropriate and will result in a
state as if the transactions were serially executed, yet
will achieve efficiency due to concurrent execution.
8/11/2019 Lec 1 Transaction Processing Concepts
56/87
Serializability of Schedules (Cont.)
Serial and Nonserial schedule
A schedule S is serial if, for every
transaction T participating in the schedule,
all the operations of T are executedconsecutively in the schedule; otherwise,
the schedule is called nonserial
Serializable scheduleA schedule S of n transactions is
serializable if it is equivalent to some
serial schedule of the same n transactions56
8/11/2019 Lec 1 Transaction Processing Concepts
57/87
Serial Schedule
We consider transactions to be independent, so
serial schedule is correct
Based on C property in ACID
Furthermore, it does not matter whichtransaction is executed first, as long as every
transaction is executed in its entirety, from
beginning to end
Example
Assume X=90, Y=90, N=3, M=2, then result of
schedule S is X=89 and Y= 93
Same result if we start with T257
Why Do We Interleave Transactions?
8/11/2019 Lec 1 Transaction Processing Concepts
58/87
Why Do We Interleave Transactions?
T1read_item(X);X:=X-N;
write_item(X);
read_item(Y);Y:=Y+N;
write_item(Y);
58
T2
read_item(X):
X:=X+M;write_item(X);
Could be a long wait
S is a serial scheduleno interleaving!
Schedule S
Another Sched le
8/11/2019 Lec 1 Transaction Processing Concepts
59/87
Another Schedule
T1read_item(X);
X:=X-N;
write_item(X);
read_item(Y);
Y:=Y+N;write_item(Y);
59
T2
read_item(X):
X:=X+M;
write_item(X);
S is a non-serial schedule
T2 will be done faster but is the result correct?
Schedule S
8/11/2019 Lec 1 Transaction Processing Concepts
60/87
Concurrent Executions
Serial execution is by far simplest methodto execute transactions
No extra work ensuring consistency
Inefficient! Reasons for concurrency:
Increased throughput
Reduces average response time Need concept of correct concurrent
execution
Using same X, Y, N, M values as before,
result of S is X=92 and Y=93 (not correct) 60
8/11/2019 Lec 1 Transaction Processing Concepts
61/87
Yet Another ScheduleT1
read_item(X);
X:=X-N;
write_item(X);
read_item(Y);
Y:=Y+N;write_item(Y);
61
T2
read_item(X):
X:=X+M;
write_item(X);
S is a non-serial schedule
Produces same result as serial schedule S
Schedule S
Serializability
8/11/2019 Lec 1 Transaction Processing Concepts
62/87
Serializability Assumption:Every serial schedule is correct
Goal: Find non-serialschedules which are alsocorrect
A schedule Sof ntransactions is serializable if it
is equivalent to some serial schedule of the
same ntransactions
When are two schedules equivalent?
Option 1: They lead to same result (result
equivalent) Option 2: The order of any two conflicting
operations is the same (conflict equivalent)
62
8/11/2019 Lec 1 Transaction Processing Concepts
63/87
Result Equivalent Schedules Two schedules are result equivalent if they
produce the same final state of the database
Problem: May produce same result by accident!
63
S1read_item(X);
X:=X+10;
write_item(X);
S2read_item(X);
X:=X*1.1;
write_item(X);
Schedules S1 and S2 are result equivalent for X=100 but not in general
8/11/2019 Lec 1 Transaction Processing Concepts
64/87
Conflict Equivalent Schedules
Two schedules are conflict equivalent, if
the order of any two conflicting operations
is the same in both schedules
64
8/11/2019 Lec 1 Transaction Processing Concepts
65/87
Conflict Equivalence
T1read_item(A);
write_item(A);
read_item(B);
write_item(B);
65
T2
read_item(A):
write_item(A);
read_item(B);
write_item(B);
order mattersorder doesnt matter
order matters
Serial Schedule S1
order
doesnt matter
C fli t E i l
8/11/2019 Lec 1 Transaction Processing Concepts
66/87
Conflict EquivalenceT1
read_item(A);
read_item(B);
write_item(A);
write_item(B);
66
T2
read_item(A):
write_item(A);
read_item(B);
write_item(B);S1 and S1 are conflict equivalent
(S1 produces the same result as S1)
Schedule S1
same order as in S1
same order as in S1
C fli t E i l
8/11/2019 Lec 1 Transaction Processing Concepts
67/87
Conflict EquivalenceT1
read_item(A);
write_item(A);
read_item(B);
write_item(B);
67
T2
read_item(A):
write_item(A);
read_item(B);
write_item(B);
Schedule S1 is not conflict equivalent to S1
(produces a different result than S1)
Schedule S1
different order than in S1
different order than in S1
Conflict Serializable
8/11/2019 Lec 1 Transaction Processing Concepts
68/87
Conflict Serializable Schedule S is conflict serializable if it is
conflict equivalent to some serialschedule S We can reorder the non-conflictingoperations to
improve efficiency
Non-conflicting operations:
Reads and writes from same transaction
Reads from different transactions
Reads and writes from different transactions on
different data items
Conflicting operations:
Reads and writes from different transactions on
same data item
68
Example
8/11/2019 Lec 1 Transaction Processing Concepts
69/87
Example
T1read_item(X);
X:=X-N;
write_item(X);
read_item(Y);
Y:=Y+N;
write_item(Y);
69
T2
read_item(X);
X:=X+M;
write_item(X);
T1read_item(X);
X:=X-N;
write_item(X);
read_item(Y);
Y:=Y+N;
write_item(Y);
T2
read_item(X);X:=X+M;
write_item(X);
Schedule A Schedule B
B is conflict equivalent to A B is serializable
Test for Serializability
8/11/2019 Lec 1 Transaction Processing Concepts
70/87
Test for Serializability Construct a directed graph,precedence graph,G = (V,
E)
V: set of all transactions participating in schedule
E: set of edges TiTjfor which one of the following holds:
Tiexecutes a write_item(X) before Tjexecutes read_item(X)
Tiexecutes a read_item(X) before Tjexecutes write_item(X)
Tiexecutes a write_item(X) before Tjexecutes write_item(X) An edge TiTj means that in any serial schedule
equivalent to S, Timust come before Tj
If G has a cycle, than S is not conflict serializable
If not, use topological sort to obtain serialiazableschedule (linear order consistent with precedence
order of graph)
70
8/11/2019 Lec 1 Transaction Processing Concepts
71/87
Sample Schedule ST1
read_item(X);
write_item(X);
read_item(Y);
write_item(Y);
71
T2
read_item(Z);
read_item(Y);
write_item(Y);
read_item(X);
write_item(X);
T3
read_item(Y);read_item(Z);
write_item(Y);write_item(Z);
8/11/2019 Lec 1 Transaction Processing Concepts
72/87
Precedence Graph for S
72
T1 T2
T3
X,Y
Y Y,Z
Equivalent Serial Schedule:
T3T1T2
(precedence order)
no cyclesS is serializable
FIGURE 17 8 (continued)
8/11/2019 Lec 1 Transaction Processing Concepts
73/87
Chapter 17-73
FIGURE 17.8 (continued)
Another example of serializability testing. (b) Schedule E.
Characterizing Schedules
8/11/2019 Lec 1 Transaction Processing Concepts
74/87
Characterizing Schedules
based on Serializability Being serializable is not the same as
being serial
Being serializable implies that the
schedule is a correct schedule.
It will leave the database in a consistent
state.
The interleaving is appropriate and will result
in a state as if the transactions were serially
executed, yet will achieve efficiency due to
concurrent execution. 74
Characterizing Schedules
8/11/2019 Lec 1 Transaction Processing Concepts
75/87
Characterizing Schedules
based on Serializability
Serializability is hard to check.
Interleaving of operations occurs in an
operating system through some scheduler Difficult to determine before hand how the
operations in a schedule will be interleaved.
75
Characterizing Schedules
8/11/2019 Lec 1 Transaction Processing Concepts
76/87
Characterizing Schedules
based on SerializabilityPractical approach: Come up with methods (protocols) to ensure
serializability.
Its not possible to determine when a schedule begins
and when it ends. Hence, we reduce the problem ofchecking the whole schedule to checking only a
committed projectof the schedule (i.e. operations from
only the committed transactions.)
Current approach used in most DBMSs: Concurrency control techniques
Examples
Two-phase locking technique
Timestamp ordering technique
76
Characterizing Schedules
8/11/2019 Lec 1 Transaction Processing Concepts
77/87
Chapter 17-77
Characterizing Schedules
based on Serializability (4)
Serializability is hard to check.
Interleaving of operations occurs in an
operating system through some scheduler Difficult to determine beforehand how the
operations in a schedule will be interleaved.
Characterizing Schedules
8/11/2019 Lec 1 Transaction Processing Concepts
78/87
Characterizing Schedules
based on Serializability View equivalence: A less restrictive definition of
equivalence of schedules
View serializability Definition of serializability based on view equivalence.
A schedule is view serializable if it is view equivalent to
a serial schedule.
78
Characterizing Schedules
8/11/2019 Lec 1 Transaction Processing Concepts
79/87
Characterizing Schedules
based on SerializabilityTwo schedules are said to be view equivalentif the following three
conditions hold:
1. The same set of transactions participates in S and S, and S and S
include the same operations of those transactions.
2. For any operation Ri(X) of Tiin S, if the value of X read by theoperation has been written by an operation Wj(X) of Tj(or if it is the
original value of X before the schedule started), the same condition
must hold for the value of X read by operation Ri(X) of Tiin S.
3. If the operation Wk(Y) of Tkis the last operation to write item Y in S,
then Wk(Y) of Tkmust also be the last operation to write item Y inS.
79
Characterizing Schedules
8/11/2019 Lec 1 Transaction Processing Concepts
80/87
Characterizing Schedules
based on SerializabilityThe premise behind view equivalence:
As long as each read operation of a transaction reads
the result of the same write operationin both schedules,
the write operations of each transaction must produce
the same results.
The view:the read operations are said to see the the
same viewin both schedules.
80
Characterizing Schedules
8/11/2019 Lec 1 Transaction Processing Concepts
81/87
Characterizing Schedules
based on SerializabilityRelationship between view and conflict equivalence:
The two are same under constrained write assumption
which assumes that if T writes X, it is constrained by the
value of X it read; i.e., new X = f(old X)
Conflict serializability is stricterthan view serializability.
With unconstrained write (or blind write), a schedule that
is view serializable is not necessarily conflict serializable.
Any conflict serializable schedule is also view
serializable, but not vice versa.
81
Characterizing Schedules
8/11/2019 Lec 1 Transaction Processing Concepts
82/87
Characterizing Schedules
based on SerializabilityRelationship between view and conflict equivalence
(cont):
Consider the following schedule of three transactions
T1: r1(X), w1(X); T2: w2(X); and T3: w3(X):
Schedule Sa: r1(X); w2(X); w1(X); w3(X); c1; c2; c3;
In Sa, the operations w2(X) and w3(X) are blind writes, since T1 and
T3 do not read the value of X.
Sa is view serializable, since it is view equivalent to the serial
schedule T1, T2, T3. However, Sa is not conflict serializable, since
it is not conflict equivalent to any serial schedule.
82
FIGURE 17.7Constructing the precedence graphs for schedulesA and Dfrom Figure 17.5 to
t t f fli t i li bilit ( ) P d h f i l h d l A (b)
8/11/2019 Lec 1 Transaction Processing Concepts
83/87
Chapter 17-83
test for conflict serializability. (a) Precedence graph for serial scheduleA. (b)
Precedence graph for serial schedule B. (c) Precedence graph for schedule C
(not serializable). (d) Precedence graph for schedule D(serializable, equivalent
to scheduleA).
T ti S t i SQL
8/11/2019 Lec 1 Transaction Processing Concepts
84/87
Transaction Support in SQL A single SQL statement is always considered
to be atomic There is no explicit Begin_Transaction
statement
SET TRANSACTIONstatement in SQL2 sets
the characteristics of a transaction Access mode
READ only or READ-WRITE
Diagnostic area size
Indicates the number of conditions that can be heldsimultaneously in the diagnostic area.
Isolation level READ UNCOMMITTED, READ COMMITTED,
REPEATABLE READ, SERIALIZABLE
84
8/11/2019 Lec 1 Transaction Processing Concepts
85/87
85
Isolation Level
Type of Violation
Dirty
READ
Non-
Repeatable
READ
Phantom
READ
UNCOMMITTEDYes Yes Yes
READ
COMMITTEDNo Yes Yes
REPEATABLE
READNo No Yes
SERIALIZABLE No No No
A Sample SQL Transaction
8/11/2019 Lec 1 Transaction Processing Concepts
86/87
A Sample SQL TransactionEXEC SQL WHENEVER SQLERROR GOTO UNDO;
EXEC SQL SET TRANSACTION
READ WRITE
DIAGONOSTIC SIZE 5
ISOLATION LEVEL SERIALIZABLE;
EXEC SQL INSERT INTO
EMPLOYEE(FNAME, LNAME, SSN, DNO, SALARY)VALUES (Ali, Al-Fares, 991004321, 2, 35000)
EXEC SQL UPDATE EMPLOYEE
SET SALARY = SALARY * 1.1 WHERE DNO = 2;
EXEC SQL COMMIT;
GOTO END_T;UNDO: EXEC SQL ROLLBACK;
END_T: ;
86
S
8/11/2019 Lec 1 Transaction Processing Concepts
87/87
Summary
Introduction to transaction processing Transaction and system concepts
Desirable properties of transactions
Schedules and recoverability Serializability of schedules
Transaction support in SQL
Thank you