Advanced Database Systems

(CoSc2042)

Transaction Management
Chapter Contents

 Introduction to Transaction Processing

 Desirable Properties of Transactions
 Transaction States
 Schedules based on Recoverability
 Schedules based on Serializability
 Transaction Support in SQL

2
 Introduction to Transaction Processing
 Single-user System:
 At most one user at a time can use the database
management system.
 Eg. Personal computer system
 Multi-user System:
 Many users can access the DBMS concurrently.
 Eg. Air line reservation, Bank and the like system are
operated by many users who submit transaction
concurrently to the system
 This is achieved by multiprogramming, which allows the
computer to execute multiple programs /processes at the
same time.

3
Introduction to Transaction Processing
 Transaction
 Logical unit of database processing that includes one or more
access operations (read -retrieval, write - insert or update,
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 an application program.

 Transaction boundaries:
 Any single transaction in an application program is bounded
with Begin and End statements.
 Application program may contain several transactions
separated by the Begin and End transaction boundaries.
4
Simple Model of Database

 A database is a collection of named data items

 Granularity of data - a field, a record , or a whole disk
block that measure the size of the data item
 Basic operations that a transaction can perform 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
Simple Model of Database…
 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
Simple Model of Database…
 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 DBMS maintains a number of buffers in the main
memory that holds database disk blocks which contains the
data base items being processed.
 Some buffer management policy is used to choose for
replacement but if the chosen buffer has been modified, it
must be written back to disk before it is used.
7
 Two sample transactions

(a) Transaction T1
(b) Transaction T2

8
Transaction Processing - Concurrency

 Interleaved processing:
 Concurrent execution of processes is interleaved in a single
CPU using for example, round robin algorithm
Advantages:
 Keeps the CPU busy when the process requires I/O by
switching to execute another process rather than
remaining idle during I/O time and hence this will increase
system throughput (average number of transactions
completed within a given time)
 Prevents long process from delaying other processes
(minimize unpredictable delay in the response time).
 Parallel processing:
 If Processes are concurrently executed in multiple CPUs.

9
10
Why Concurrency Control is needed ?
 Lost Update Problem
 Temporary Update (Dirty Read) Problem
 Incorrect Summary Problem

11
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.

12
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.

Fig 2: Temporal update problem

13
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.

14
Desirable Properties of Transactions
ACID Properties:
 Atomicity: A transaction is an atomic unit of processing; it
is either performed entirely 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, 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.

15
Transaction States
 A transaction is an atomic unit of work that is either completed
in its entirety or not done at all.
 For recovery purposes, the system needs to keep track of when
the transaction starts, terminates, and commits or aborts.
 Transaction states:
 Active state: indicates the beginning of a transaction
execution
 Partially committed state: shows the end of read/write
operation but this will not ensure permanent modification on
the database
 Committed state: ensures that all the changes done on a
record by a transition were done persistently
 Failed state: happens when a transaction is aborted during
its active state or if one of the rechecking is fails
 Terminated State: corresponds to the transaction leaving the
system
16
Transaction States…

17
What causes a Transaction to fail ?

1. Computer failure (system crash):

 A hardware or software error occurs in the computer system
during transaction execution.
 If the hardware crashes, the contents of the computer’s internal
memory may be lost.
2. 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.
 In addition, the user may interrupt the transaction during its
execution.
3. Exception conditions detected by the transaction:
 Certain conditions forces cancellation of the transaction.
 Data for the transaction may not be found, insufficient account
balance in a banking database, may cause a transaction, such
as a fund withdrawal from that account, to be canceled.

18
What causes a Transaction 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.
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,
overwriting disks or tapes by mistake.

19
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.
 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 (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.

 Recovery techniques will be discussed in Next Chapter.

20
Schedules based on Recoverability
 Transaction schedule or history:
 When transactions are executing concurrently in
an interleaved fashion, the order of execution of
operations from the various transactions forms
what is known as a transaction schedule
(history).
 A schedule S of n transactions T1, T2, …, Tn:
 It 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 T1 in S must appear in the same
order in which they occur in Ti.
 Note, that operations from other transactions Tj
can be interleaved with the operations of Ti in S.
21
Schedules based on Recoverability…
For example
Sa : r1(X);w1(X);r2(X);w2(X);r1(Y);a1;

22
Schedules based on Recoverability…
 Two operations in a schedule are side to be
conflict if they satisfy all the three of the following
conditions.
 They belongs to different transaction
 They access the same data item X
 At least one of the operation is a write_Item(X)

Eg. Sa: r1(X); r2(x); w1(X); r1(Y); W2(X); W1(Y);

 r1(X) and w2(X) Conflict operations
 r2(X) and w1(X);
 W1(X) and w2(X)

No Conflict
 r1(X) and r2(X)
 W2(X) and w1(Y)

23
Schedules based on Recoverability…
Classified into four types
 Recoverable schedule:
 One where no committed transaction needs to be rolled back.
 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.
Examples,
 Sc: r1(X); w1(X); r2(X); r1(Y);w2(x);c2;a1; not recoverable
 Sd: r1(X); w1(X); r2(X); r1(Y); w2(X);w1(Y); c1; c2; Recoverable
 Se: r1(X); w1(X); r2(X); r1(Y); w2(x) ; w1(Y); a1; a2;
 Cascaded schedule:
 A schedule in which uncommitted transactions that read an item from a
failed transaction must be rolled back.
 Se: r1(X); w1(X); r2(X); r1(Y); w2(x) ; w1(Y); a1; a2;
 Cascadeless schedule:
 One where every transaction reads only the items that are written by
committed transactions.
 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);
24
Schedules based on Serializability
 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 non serial schedule.
 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 non serial schedule S is serializable equivalent, if it is
correct to the result of one of the serial schedule .
 For Example consider diagram on next slide.

25
Fig 4:

26
Schedules based on Serializability …
 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
i. Conflict equivalent:
 Two schedules are said to be conflict equivalent if the
order of any two conflicting operations is the same in
both schedules. Eg
Not conflict equivalent
 S1: r1(x); w2(x) & S2: w2(x); r1(x)
 S1: w1(x); w2(x); & S2: w2(x); w1(x);
 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 .

27
 Conflict Equivalence
 If you can transform an interleaved schedule by
swapping consecutive non-conflicting operations
of different transactions into a serial schedule,
then the original schedule is conflict serializable.
 Example:

R(A) W(A) R(B) W(B)

R(A) W(A)
R(B)W(B)

R(A) W(A) R(B)R(B)

W(B)R(B)
W(B)W(B)

28
R(A) W(A)
R(A)R(A) W(A)R(B)W(B)
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.
 Interleaving is appropriate and will result in a
state as if the transactions were serially
executed, yet will achieve efficiency due to
concurrent execution.
 Serializability is hard to check
 Difficult to determine beforehand how the
operations in a schedule will be interleaved.

29
Practical Approach
 Come up with methods (protocols) to ensure
serializability.
 It’s not possible to determine when a schedule
begins and when it ends.
 Hence, reducing the problem of checking the
whole schedule to checking only a committed
project of the schedule (i.e. operations from
only the committed transactions.)
 Current approach used in most DBMSs:
 Use of locks with two phase locking
 Detail in chapter two.

30
 View Equivalence
 A less restrictive definition of equivalence of schedules
View serializability:
 A schedule is view serializable if it is view equivalent to a
serial schedule.
 Two schedules are said to be view equivalent if the following
three conditions hold:
 The same set of transactions participates in S and S’, and
S and S’ include the same operations of those
transactions.
 If Ti reads a value A written by Tj in S1 , it must also read
the value
S’ of A written by Tj in S2 S
 for each data object A, the transaction that perform the
T1: R(A)
final writeW(A)
on x in S1 must alsoT1: R(A),W(A)
perform the final write on
T2: A inW(A) S2 view T2: W(A)
T3: T3:
W(A) W(A)
31
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.
 Any conflict serializable schedule is also view
serializable, but not vice versa.

32
Relationship between view and conflict
equivalence…
 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
Testing for equivalent to any serial schedule.
conflict serializability: Algorithm
• Looks at only read_Item (X) & write_Item (X) operations
• Constructs a precedence graph (serialization graph) - a graph
with directed edges
• An edge is created from Ti to Tj if one of the operations in Ti
appears before a conflicting operation in Tj
• The schedule is serializable if and only if the precedence graph
33 has no cycles.
Constructing the Precedence Graphs
 FIGURE 5: Constructing the precedence graphs for schedules A and D from
Figure 4 (from slide No 26) to test for conflict serializability.
 (a) Precedence graph for serial schedule A.
 (b) Precedence graph for serial schedule B.
 (c) Precedence graph for schedule C (not serializable).
 (d) Precedence graph for schedule D (serializable, equivalent to schedule
A).

34
Another Example of Serializability Testing

35
36
37
 Summery of Schedule types

38
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
1. 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.
2. Diagnostic size n, specifies an integer value n, indicating the number
of error conditions that can be held simultaneously in the diagnostic area.

3. Isolation level can be

 READ UNCOMMITTED,
 READ COMMITTED,
 REPEATABLE READ or
 SERIALIZABLE. The default is SERIALIZABLE.
39
Transaction Support in SQL …
 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: ...
• With SERIALIZABLE: the interleaved execution of transactions will adhere to the
notion of serializability.
• However, if any transaction executes at a lower level, then serializability may be
violated.
40
Potential Problem with Lower Isolation Levels

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), 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
41
 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.
42
Possible Violation of Serializabilty
Type of Violation_______________
Isolation Dirty non repeatable
level read read phantom
_______________________________________________________
READ UNCOMMITTED yes yes yes
READ COMMITTED no yes yes
REPEATABLE READ no no yes
SERIALIZABLE no no no

43