Advanced Database Systems (SEng4122)

Lecture 1

1
Transaction Processing Concepts

 Outline
 Introduction to Transaction Processing
 Transaction and System Concepts
 Desirable Properties of Transactions
 Characterizing Schedules based on Recoverability
 Characterizing Schedules based on Serializability
 Transaction Support in SQL

2
Introduction
 Single user Vs multiuser systems
 One criterion for classifying a database system is
according to the number of users who can use the
system at the same time
 Single-User System:
A DBMS is a single user if at most one user at a time can use the
system.
 Multiuser System:
Many users can access the system concurrently.
 Concurrency
Interleaved processing:
 Concurrent execution of processes is interleaved in a single

CPU, using for example, round robin algorithm

Parallel processing:
 Processes are concurrently executed in multiple CPUs.

3
Introduction (cont…)
 A Transaction:
 Logical unit of database processing that includes one or more
access operations (read, retrieval, write, insert or update and
delete)
 A transaction (set of operations) may be stand-alone
specified in a high level language like SQL submitted
interactively, or may be embedded within a program.
 Examples include ATM transactions, credit card approvals, flight
reservations, hotel check-in, phone calls, supermarket scanning,
academic registration and billing.
 Transaction boundaries:
 One way of specifying transaction boundaries is using explicit
Begin and End transaction statements in an application
program
 An application program may contain several transactions
separated by the Begin and End transaction boundaries

4
Introduction (cont…)

Simple Model of a Database (for purposes of discussing transactions):

 A database is a collection of named data items

 Granularity of data - a field, a record , or a whole disk

block . Transaction concepts are independent of granularity

 Basic operations are read and write

 read_item(X): Reads a database item named X into a

program variable. To simplify our notation, we assume

that the program variable is also named X.
 write_item(X): Writes the value of program variable X

into the database item named X.

5
Introduction (cont…)
Read and write operations:
 Basic unit of data transfer from the disk to the computer
main memory is one block.
 In general, a data item (what is read or written) will be

the field of some record in the database, although it

may be a larger unit such as a record or even a whole
block
 read_item(X) command includes the following steps:
 Find the address of the disk block that contains item X.

 Copy that disk block into a buffer in main memory (if

that disk block is not already in some main memory

buffer).
 Copy item X from the buffer to the program variable

named X.

6
Introduction (cont…)
Read and Write Operations (cont.):
 write_item(X) command includes the following steps:
 Find the address of the disk block that contains item X.

 Copy that disk block into a buffer in main memory (if

that disk block is not already in some main memory

buffer)
 Copy item X from the program variable named X into

its correct location in the buffer.

 Store the updated block from the buffer back to disk

(either immediately or at some later point in time).

 The decision about when to store back a modified

disk block that is in main memory is handled by the

recovery manager of the DBMS in cooperation with
the underlying operating system

7
Introduction (cont…)
 Example of transactions
 (a) Transaction T1
 (b) Transaction T2

8
Introduction (cont…)
 Transactions submitted by the various users may execute
concurrently and may access and update the same
database items
 If this concurrent execution is uncontrolled, it may lead to
problems such as inconsistent database
 Why Concurrency Control is needed:
 Concurrency control is needed to respond to the effect of
the following problems on database consistency
 The Lost Update Problem
 This occurs when two transactions that access the same
database items have their operations interleaved in a
way that makes the value of some database item
incorrect since the update made by the first transaction
is not used by the second transaction.
 In other words, the update made by the fist transaction
is lost(overwritten) by the second transaction
9
Introduction (cont…)
 The Temporary Update (Dirty Read) Problem
 This occurs when one transaction updates a database

item and then the transaction fails for some reason.

 The updated item is accessed by another transaction

before it is changed back to its original value.

 The Incorrect Summary Problem
 If one transaction is calculating an aggregate

summary function on a number of records while

other transactions are updating some of these
records, the aggregate function may calculate some
values before they are updated and others after they
are updated.

10
Concurrent execution is uncontrolled:
(a) The lost update problem

11
 E.g. Account with balance A=100.
 T1 reads the account A
 T1 withdraws 10 from A
 T1 makes the update in the Database
 T2 reads the account A
 T2 adds 100 on A
 T2 makes the update in the Database
 In the above case, if done one after the other (serially) then we have no problem.
 If the execution is T1 followed by T2 then A=190
 If the execution is T2 followed by T1 then A=190
 But if they start at the same time in the following sequence:
 T1 reads the account A=100 T1 T2
Read_item(A)
 T1 withdraws 10 making the balance A=90
A=A-10
 T2 reads the account A=100 Read_item(A)
 T2 adds 100 making A=200 A=A+100
Write_item(A)
 T1 makes the update in the Database A=90
Write_item(A)
 T2 makes the update in the Database A=200

 After the successful completion of the operation the final value of A will be 200
which override the update made by the first transaction that changed the value from
100 to 90.
12
12
Lost Update problem: solution

Lost update!!
This could have been avoided if we prevent T2 from reading
until T1’s update has been completed

13
Concurrent execution is uncontrolled:
(b) The temporary update problem.

14
Example: T2 increases 100 making it 200 but then aborts the transaction
before it is committed. T1 gets 200, subtracts 10 and make it 190. But
the actual balance should be 90
T1 T2

Read_item(A)

A=A+100

Transaction T2 fails and must

Write_item(A) change the values of A back to its
old value; Meanwhile T1 has read
the temporary incorrect value of A
Read_item(A)

A=A-10

Write_item(A)

Abort

15
The temporary update problem: Example

Time T1 T2 bal(X)
t1 Begin Tx 100
t2 R(balX) 100
t3 balx=balx+100 100
t4 Begin Tx W(balx) 200
t5 R(balX) 200
t6 balx=balx-10 Rollback 200
t7 W(balx) 190
t8 Commit 190

 Temporary update!!
 Could have been avoided if we prevent T1 from reading until after
the decision to commit or rollback T2 has been made
16
Concurrent execution is uncontrolled:
(c) The incorrect summary problem.

17
The incorrect summary problem: Example
Time T5 T6 Bal(x) Bal(z) Sum
t1 Begin Tx 100 25 0
t2 Begin Tx Sum=0 100 25 0
t3 R(balX) 100 25 0
t4 balx=balx-10 R(balX) 100 25 0
t5 Sum+=balx 100 25 100
W(balx)
t6 R(balZ) 90 25 100
t7 balz=balz+10 90 25 100
t8 W(balz) 90 35 100
t9 Commit R(balz) 90 35 100
t10 Sum+=balz 90 35 135
t11 W(sum) 90 35 135
t12 commit 90 35 135
The incorrect summary problem:
• Example 2: T1 would like to add the values of A=10, B=20 and C=30. after
the values are read by T1 and before its completion, T2 updates the
value of B to be 50. at the end of the execution of the two transactions
T1 will come up with the sum of 60 while it should be 90 since B is
updated to 50
T1 T2
Sum= 0;
Read_item(A)
Sum=Sum+A
Read_item(B)
Sum=Sum+B
Read_item(B)
B=50
Read_item(C)
Sum=Sum+C

19
What causes a Transaction to fail?
1. A computer failure (system crash):
 A hardware or software error may occur in the
computer system during transaction execution. If
the hardware crashes, the contents of the
computer’s internal memory may be lost.
2. A transaction or system error:
 Some operation in the transaction may cause it to
fail, such as integer overflow or division by zero.
 Transaction failure may also occur because of
erroneous parameter values or because of a
logical programming error

20
What causes a Transaction to fail (Cont...)
3. Local errors or exception conditions detected by the
transaction:
 Certain conditions necessitate cancellation of the
transaction
 For example, data for the transaction may not
be found
 A programmed abort in the transaction causes it to
fail.
4. Concurrency control enforcement:
The concurrency control method may decide to
abort the transaction, to be restarted later, because
it violates serializability or because several
transactions are in a state of deadlock

21
What causes a Transaction to fail (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 mounting of a wrong tape by the operator.

22
Transaction and System Concepts

 Transaction states and additional operations

 Transaction states:
 Active state

 Partially committed state

 Committed state

 Failed state

 Terminated State

23
State transition diagram illustrating the
states for transaction execution

24
Transaction and System Concepts (cont…)
 Transaction operations
 For recovery purposes, the system needs to keep track of when
the transaction starts, terminates, and commits or aborts
 Recovery manager keeps track of the following operations:
 begin_transaction: This marks the beginning of transaction

execution
 read or write: These specify read or write operations on the

database items that are executed as part of a transaction

 End_transaction: This specifies that read and write transaction
operations have ended and marks the end limit of transaction
execution.
 At this point it may be necessary to check whether the

changes introduced by the transaction can be permanently

applied to the database or whether the transaction has to be
aborted because it violates concurrency control or for some
other reason.

25
Transaction and System Concepts (cont…)

 commit_transaction:
 This signals a successful end of the transaction so that

any changes (updates) executed by the transaction

can be safely committed to the database and will not
be undone.
 rollback (or abort):
 This signals that the transaction has ended

unsuccessfully, so that any changes or effects that the

transaction may have applied to the database must be
undone.

26
Transaction and System Concepts (cont…)
 The System Log

 Log or Journal: The log keeps track of all

transaction operations that affect the values of
database items
 This information is needed to permit recovery from
transaction failures
 The log is kept on disk, so it is not affected by any
type of failure except for disk or catastrophic failure.
 In addition, the log is periodically backed up to
archival storage (tape) to guard against such
catastrophic failures.

27
Transaction and System Concepts (cont…)
 The System Log (cont):

 We can use a notation T to refer to a unique

transaction-id that is generated automatically by the
system and is used to identify each transaction:
 Types of log record:
 [start_transaction,T]: Records that transaction T has

started execution.
 [write_item,T,X,old_value,new_value]: Records that

transaction T has changed the value of database

item X from old_value to new_value.

28
The System Log (cont):
 [read_item,T,X]: Records that transaction T has
read the value of database item X.
 [commit,T]: Records that transaction T has
completed successfully, and affirms that its effect
can be committed (recorded permanently) to the
database.
 [abort,T]: Records that transaction T has been
aborted.

29
Recovery using log records:
 If the system crashes, we can recover to a consistent
database state by examining the log record and using
recovery methods.
1. Because the log contains a record of every write
operation that changes the value of some database
item, it is possible to undo the effect of these write
operations of a transaction T by tracing backward
through the log and resetting all items changed by a
write operation of T to their old_values.
2. We can also redo the effect of the write operations of
a transaction T by tracing forward through the log and
setting all items changed by a write operation of T
(that did not get done permanently) to their
new_values.

30
Transaction and System Concepts (cont…)
Commit Point of a Transaction:
 Definition a Commit Point:

 A transaction T reaches its commit point when all its

operations that access the database have been executed

successfully and the effect of all the transaction
operations on the database has been recorded in the log
 Beyond the commit point, the transaction is said to be

committed, and its effect is assumed to be permanently

recorded in the database.
 The transaction then writes a commit record
[commit,T] in to the log

31
Transaction and System Concepts (cont…)
 Undoing transactions
 If a system failure occurs, we search back in the log for

all transactions T that have written a

[start_transaction,T] entry into the log but no commit
entry [commit,T] record yet
 These transactions have to be rolled back to undo

their effects on the database during recovery

process

32
Transaction and System Concepts (cont…)

Commit Point of a Transaction (cont):

 Redoing transactions:

 Transactions that have written their commit entry in the

log must also have recorded all their write operations in

the log; otherwise they would not be committed, so
their effect on the database can be redone from the log
entries. (Notice that the log file must be kept on disk.
 At the time of a system crash, only the log entries that

have been written back to disk are considered in the

recovery process because the contents of main
memory may be lost.

33
Desirable Properties of Transactions
 Transaction should posses several properties. They are
often called the ACID properties and should be enforced by
the concurrency control and recovery methods of the DBMS.
ACID properties:
 Atomicity: A transaction is an atomic unit of processing; it is
either performed in its entirety or not performed at all.
 Consistency preservation: A correct execution of the
transaction must take the database from one consistent
state to another.
 Isolation: A transaction should not make its updates visible
to other transactions until it is committed; this property, when
enforced strictly, solves the temporary update problem and
makes cascading rollbacks of transactions unnecessary
 Durability or permanency: Once a transaction changes the
database and the changes are committed, these changes
must never be lost because of subsequent failure.

34
Example:
 Suppose that Ti is a transaction that transfer 200 birr from account
CA2090( which is 5,000 Birr) to SB2359(which is 3,500 birr) as follows
 Read(CA2090)
 CA2090= CA2090-200
 Write(CA2090)
 Read(SB2359)
 SB2359= SB2359+200
 Write(SB2359)
 Atomicity- either all or none of the above operation will be done – this is
materialized by transaction management component of DBMS
 Consistency-the sum of CA2090 and SB2359 be unchanged by the
execution of Ti i.e 8500- this is the responsibility of application
programmer who codes the transaction
 Isolation- when several transaction are being processed concurrently
on a data item they may create many inconsistent problems. So
handling such case is the responsibility of Concurrency control
component of the DBMS
 Durability - once Ti writes its update this will remain there when the
database restarted from failure . This is the responsibility of recovery
management components of the DBMS 35
35
Schedules
 Schedule (or history) of transaction
 When transactions are executing concurrently in an interleaved

fashion, the order of execution of operations from the various

transactions form what is known as a transaction schedule (or
history)
 A schedule (or history) S of n transactions T1, T2,.. ,Tn:
 is an ordering of the operations of the transactions subject to the
constraint that, for each transaction Ti that participates in S, the
operations of Ti in S must appear in the same order in which they
occur in Ti.
 Note, however, that operations from other transactions Tj can

be interleaved with the operations of Ti in S.

36
Schedules (cont…)
 A shorthand notation for describing a schedule uses the
symbols :
 r : for read_item operations ,
 w: write_item,
 c: commit and
 a: abort
 Transaction numbers are appended as subscript to each
operation in the schedule
 The database item X that is read or written follows the r
and w operations in parenthesis
 Example:

Sa: r1(X),r2(x),w1(x), r1(Y),w2(x);w1(Y)

Sb: r1(X),w1(x),r2(x), w2(x), r1(Y),a1

37
Conflicting operations
 Two operations in a schedule are said to conflict if they
satisfy all three of the following conditions:
 They belong to different transactions

 They access the same item X

 At least one of the operations is a write_item(X)

 For example, in a schedule Sa, the operations

 r1(x) and w2(X) conflict, as do the operations r2(X) and

w1(X) and the operations w1(X) and w2(X)

38
Non conflicting operations
 The operations r1(x) and r2(x) do not conflict since both of
them are read operations
 r1(x) and w1(x) do not conflict because they belong to the
same transaction
 W2(x) and w1(y) do not conflict since they operate on
distinct data items x and y

39
Complete schedules
 A schedule S of n transactions T1, T2, ……..,Tn is
said to be a complete schedule if the following
conditions hold:
1. The operations in S are exactly those operations
in T1, T2, …Tn including a commit or abort
operations as the last operation for each
transaction in the schedule
2. For any pair of operations from the same
transaction Ti, their order of appearance in S is the
same as their order of appearance in T
3. For any two conflicting operations, one of the two
must occur before the other in the schedule
(theoretically, it is not necessary to determine an
order b/n pair of non conflicting operations)

40
Complete schedules (cont…)
 Condition (3) above allows for two non conflicting
operations to occur in the schedule without defining
which occurs first leading to the definition of partial
order of the operations in n tractions

41
Complete schedules (cont…)
 In general, it is difficult to encounter complete
schedules in a transaction processing system,
because new transactions are continually being
submitted to the system
 Hence, it is useful to define committed projection
C(S) of a schedule S, which includes only the
operations in S that belong to committed
transactions – that is transactions Ti whose
commit operation ci is in S

42
Characterizing Schedules based on Recoverability

43
Characterizing Schedules based on
Recoverability
 Schedules classified based on recoverability:
 Recoverable schedule:

 Once a transaction T is committed, it should never be necessary

to rollback T
 The schedules that theoretically meet this criterion are called
recoverable and those that do not are non recoverable
 A schedule S is recoverable if no transaction T in S commits until
all transactions T’ that have written an item that T reads have
committed
 A transaction T2 reads from Transaction T1 in a schedule S if
some item X is first written by T1 and latter read by T2
 In addition, T1 should not have been aborted before T2

reads item X and there should be no transaction that write

X after T1 writes it and before T2 reads X
44
Characterizing Schedules based on
Recoverability
 Consider the schedule given Sa’ where two commit
operations have been added to S a :
 Sa’ : r1(X),r2(x),w1(x), r1(Y),w2(x);c2;w1(Y);c1
 Sa’ is recoverable despite it suffers from lost update

problem
 However, consider the two schedules S c and Sd below:

Sc:r1(x);w1(x);r2(x);r1(y);w2(x);c2;a1

Sc is not recoverable because T2 reads X from T1 and
then T2 commits before T1 commits.

If T1 aborts after the c2 operations in Sc, then the
value of x that T2 read is no longer valid and T2 must
be aborted after it had been committed, leading to a
schedule that is not recoverable

45
Recoverability (cont…)
 For the above schedule to be recoverable, the c2
operation in Sc must be postponed until after T1
commits as shown in Sd


 Sd:r1(x);w1(x);r2(x);r1(y);w2(x);w1(y);c1;c2
Recoverable

46
Recoverability (cont…)
 If T1 aborts instead of committing, then T2 should also abort
as shown in Se because the X it read is no longer valid
Se:r1(x);w1(x);r2(x);r1(y);w2(x);w1(y);a1;a2 Recoverable

 Cascadeless schedule Vs cascading rollback

 Schedules requiring cascaded rollback:
 A schedule in which uncommitted transactions that read an

item from a failed transaction must be rolled back

 Cascadeless schedule:
 One where every transaction reads only the items that are

written by committed transactions

47
Cascadeless schedule:
One where every transaction reads only the items that are written by
committed transactions. Eg.
Sf: r1(X); w1(X); r1(Y); c1; r2(X); w2(X);w1(Y); c2;
Strict Schedules:
A schedule in which a transaction can neither read or write an item X
until the last transaction that wrote X has committed/aborted.
Eg. Sg: w1(X,5) ; c1; w2(x,8);

48
Characterizing Schedules based on Serializability

49
Characterizing Schedules based on
Serializability
 Serial schedule:
 A schedule S is serial if, for every transaction T
participating in the schedule, all the operations of T
are executed consecutively in the schedule
 Otherwise, the schedule is called non serial
schedule.
 Serializable schedule:
 A schedule S is serializable if it is equivalent to
some serial schedule of the same n transactions

50
Characterizing Schedules based on
Serializability (cont….)
 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.

51
Characterizing Schedules based on
Serializability (cont…)
 It’s difficult to determine when a schedule begins
and when it ends.
 Hence, we reduce the problem of checking the
whole schedule to checking only a committed
projection of the schedule (i.e. operations from
only the committed transactions.)
 Current approach used in most DBMSs:
 Use of locks with two phase locking

52
– The concept of Serializable of schedule is used to identify which
schedules are correct when concurrent transactions executions have
interleaving of their operations in the schedule
 Serial schedule:
 A schedule S is serial if, for every transaction T participating in the

schedule, all the operations of T are executed consecutively in the

schedule. Otherwise, the schedule is called nonserial schedule.
 For example, in the banking example suppose there are two

transaction where one transaction calculate the interest on the

account and another deposit some money into the account. hence
the order of execution is important for the final result
 Serializable schedule:
 a schedule whose effect on any consistent database
instance is identical to that of some complete serial
schedule over the set of committed transactions in S.
 A nonserial schedule S is serializable is equivalent to say that it is
correct to the result of one of the serial schedule .Example,

53
54
Characterizing Schedules based on
Serializability (cont….)
 Result equivalent:
 Two schedules are called result equivalent if they

produce the same final state of the database.

 Two types of equivalent schedule: Conflict and view

 Conflict equivalent:
 Two schedules are said to be conflict equivalent if

the order of any two conflicting operations is the

same in both schedules. E.G
 S1: r1(x); w2(x) & S2: w2(x); r1(x) Not conflict equivalen
 S1: w1(x); w2(x); & S2: w2(x); w1(x);

55
cont..
 Conflict serializable:
 A schedule S is said to be conflict serializable if it is

conflict equivalent to some serial schedule S’.

 Every conflict serializable schedule is serializable .

56
 Two schedules are said to be view equivalent if 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. If Ti reads a value A written by Tj in S1 , it must also
read the value of A written by Tj in S2
3. for each data object A, the transaction that perform
the final write on x in S1 must also perform the final
write on A in S2

S’ S
T1: R(A) W(A) T1: R(A),W(A)
T2: W(A) view T2: W(A)
T3: W(A) T3: W(A)

57
 Relationship 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 stricter than view serializability.
With unconstrained write (or blind write), a schedule that is
view serializable is not necessarily conflict serializable.

58
  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.

59
Determining conflict serializability
 To determine serializability, first identify the pair of
conflicting operations and check if their order is preserved in
one of the possible serial schedules
 schedule A:
 r1(x);w1(x),r1(y);w1(y);r2(x);w2(x)- serial schedule
 schedule B:
 r2(x);w2(x); r1(x);w1(x),r1(y);w1(y)- serial schedule
 schedule C:
 r1(x);r2(x);w1(x);w2(x),w1(y)- (not serializable).
 Schedule D:
 r1(x);w1(x);r2(x);w2(x);r1(y);w1(y)-(serializable, equivalent to
schedule A).

60
Serializability (cont…)
Testing for conflict serializability with precedence graphs:
Algorithm
 For each transaction Ti participating in Schedule S, create a node

labeled Ti in the precedence graph

 For each case in S where Tj executes read_item(x) after Ti

executes a write_item(x) create an edge (Ti Tj) in the

precedence graph
 For each case in S where Tj executes write_item(x) after Ti

executes a read_item(x) create an edge (Ti Tj) in the

precedence graph
 For each case in S where Tj executes write_item(x) after Ti

executes a write_item(x) create an edge (Ti Tj) in the

precedence graph
 The schedule is serializable if and only if the precedence graph

has no cycles.
61
Testing serializability with Precedence Graphs

Serial
Serial

Not Serializable

Serializable

62
 Transaction Support in SQL
 A single SQL statement is always considered to be atomic.
 Either the statement completes execution without error or it fails and
leaves the database unchanged.
 Every transaction has three characteristics: Access mode, Diagnostic size
and isolation
i. Access mode:
 READ ONLY or READ WRITE
 If the access mode is Read ONLY , INSERT, DELET ,
UPDATE & CREATE commands cannot be executed on the
data base
 The default is READ WRITE unless the isolation level of
READ UNCOMITTED is specified, in which case READ
ONLY is assumed.
ii. Diagnostic size n, specifies an integer value n, indicating the
number of error conditions that can be held simultaneously in the
diagnostic area.
iii. Isolation level can be
 READ UNCOMMITTED,
 READ COMMITTED,
 REPEATABLE READ or
 SERIALIZABLE. The default is SERIALIZABLE.
63
 With SQL, there is no explicit Begin Transaction
statement.
 Transaction initiation is done implicitly when
particular SQL statements are encountered.
 Every transaction must have an explicit end
statement, which is either a COMMIT or
ROLLBACK.

64
 Sample SQL transaction:
EXEC SQL whenever sqlerror go to UNDO;
EXEC SQL SET TRANSACTION
READ WRITE
DIAGNOSTICS SIZE 5
ISOLATION LEVEL SERIALIZABLE;
EXEC SQL INSERT
INTO EMPLOYEE (FNAME, LNAME, SSN, DNO, SALARY)
VALUES ('Robert','Smith','991004321',2,35000);
EXEC SQL UPDATE EMPLOYEE
SET SALARY = SALARY * 1.1
WHERE DNO = 2;
EXEC SQL COMMIT;
GOTO THE_END;
UNDO: EXEC SQL ROLLBACK;
THE_END: ...

65
  Potential problem with lower isolation levels: Four types

i. Unrepeatable Reads: RW Conflicts

• a transaction T2 could change the value of an object A that has been
read by a transaction T1, while T1 is still in progress.
• If T1 tries to read the value a again it will get a different value
T1: R(A), R(A), W(A), C
T2: R(A), W(A), C
ii. Reading Uncommitted Data ( “dirty reads”): WR Conflicts
• a transaction T2 could read a database object A that has been
modified by another transaction T1, which has not yet
committed.
T1: R(A), W(A), R(B), W(B), Abort
T2: R(A), W(A), C

66
 iii. Overwriting Uncommitted Data: WW Conflicts
• A transaction T2 could overwrite the value of an object A,
which has already been modified by a transaction T1,
while T1 is still in progress.
T1: W(A), W(B), C
T2: W(A), W(B), C
iv. Phantoms:
 New rows being read using the same read with a condition.
 A transaction T1 may read a set of rows from a table,
perhaps based on some condition specified in the SQL
WHERE clause.
 Now suppose that a transaction T2 inserts a new row that
also satisfies the WHERE clause condition of T1, into the
table used by T1.
 If T1 is repeated, then T1 will see a row that previously did
not exist, called a phantom. 67
Transaction Support in SQL
 Possible violation of serializabilty:

Type of Violation

Isolation Dirty nonrepeatable

level read read phantom
_______________________________________________________
READ UNCOMMITTED yes yes yes
READ COMMITTED no yes yes
REPEATABLE READ no no yes
SERIALIZABLE no no no

68
Summary
 Transaction and System Concepts
 Desirable Properties of Transactions
 Characterizing Schedules based on Recoverability
 Characterizing Schedules based on Serializability
 Transaction Support in SQL

69
Thank You

70