Lab 14

Transaction Management
FACULTY OF CS & IT
UNIVERSITY OF GUJRAT
Supporting Concurrent Access
 Many applications require a lot of users to access the
data simultaneously (e.g. airline booking systems)
 Uncontrolled simultaneous access can result in
confusion, so some controlling mechanism is required
 A transaction
is a logical unit of work which takes the
DB from one consistent state to another, i.e. obeying
constraints
 Itwill probably be made up of smaller operations which
temporarily cause inconsistency
What is a Transaction?
 Database transactions are logical units of work which
must ALL be performed to maintain data integrity
 A logical unit of work on a database may be a:
◦ An entire program
◦ A portion of a program
◦ A single command
 The entire series of steps necessary to accomplish a
logical unit of work
 Successful transactions change the database from one
CONSISTENT STATE to another CONSISTENT
STATE
(One where all data integrity constraints are satisfied)
Example Transaction
Database transactions are logical units of

work which must ALL be performed to
maintain data integrity
E.g. Transfer money from one account to
another
◦ UPDATE Account SET balance = balance – 100
WHERE accountNo = 123;
◦ UPDATE Account SET balance = balance + 100
WHERE accountNo = 124;
Example Transaction
5
Transaction Support
Can have one of two outcomes:

◦ Success - transaction commits and database reaches a new
consistent state.
◦ Failure - transaction aborts, and database must be restored
to consistent state before it started.
◦ Such a transaction is rolled back or undone.
Committed transaction cannot be aborted.
Aborted transaction that is rolled back can
be restarted later.
6
State Transition Diagram for Transaction
7
4 Properties of a Transaction
Atomicity – All or Nothing
All parts of the transaction must be
completed and committed If not, it must
be aborted and rolled back.
The entire transaction is treated as a
single, indivisible unit of work which
must be performed completely or not at
all.
Atomicity
All or nothing. All steps that are part of a

transaction are completed, or none are.
S= $300 S=$300-100 C=$800+100 S=$200

C=$800 Begin Commit C=$900
S= $300 S=$300-100 S=$300

C=$800 C=$800
Begin
Rollback
9
Commit and Rollback
Two key operations to ensure atomicity:
◦ Commit signals successful end of transactions.
Updates made by current transaction can be
“committed” or made permanent in database.
◦ Rollback signals unsuccessful end of transactions.
All the updates made by current transaction must be
“rolled back” or undone.
10
Consistency
Each user is responsible to ensure that their
transaction (if executed by itself) would leave
the database in a consistent state
 If an operation is executed that violates the
database’s integrity constraints, the entire
transaction will be rolled back.
A successful transaction takes the database from
one state that is consistent with the rules to
another state that is also consistent with the
rules.
Consistency Example
Loss of T2’s update avoided by preventing

T1 from reading balx until after update.
12
Isolation
The final effects of multiple simultaneous
transactions must be the same as if they
were executed one right after the other.
Data used within a transaction cannot be
used by another transaction until the first
transaction is completed. (or it must
appear that this happened!). The partial
effects of incomplete transactions should
not be visible to other transactions.
Isolation Example

14
Durability
If a transaction has been committed, the
DBMS must ensure that its effects are
permanently recorded in the database
(even if the system crashes)
So the effects of a committed transaction
are not lost in a failure.
Need for Concurrency Control
Three examples of potential problems
caused by concurrency:
◦ Lost update problem.
◦ Uncommitted dependency problem.
◦ Inconsistent analysis problem.
16
Lost Update Problem
Successfully completed update is
overridden by another user.
T1 withdrawing £10 from an account with
balx, initially £100.
T2 depositing £100 into same account.
Serially, final balance would be £190.
17
Lost Update Problem

18
Uncommitted Dependency Problem
Occurs when one transaction can see
intermediate results of another transaction
before it has committed.
T updates bal to £200 but it aborts, so bal
4 x x
should be back at original value of £100.
T has read new value of bal (£200) and uses
3 x
value as basis of £10 reduction, giving a new
balance of £190, instead of £90.
19
Uncommitted Dependency Problem
 Problem avoided by preventing T3 from reading

balx until after T4 commits or aborts.
20
Inconsistent Analysis Problem
 Occurs when transaction reads several values but
second transaction updates some of them during
execution of first.
 Sometimes referred to as dirty read or unrepeatable
read.
 T is totaling balances of account x (£100), account
6
y (£50), and account z (£25).
 Meantime, T has transferred £10 from bal to
5 x
balz, so T6 now has wrong result (£10 too high).
21
Inconsistent Analysis Problem
 Problem avoided by preventing T6 from reading balx

and balz until after T5 completed updates.
22
Transaction Management with SQL
 SQL Statements  Commit / Rollback
 When a transaction sequence is initiated it must continue
through all succeeding SQL statements until:
1. A Commit Statement is Reached
2. A Rollback Statement is Reached
3. The End of the Program is Reached (Commit)
4. The Program is Abnormally Terminated (Rollback)

Lab 14

Uploaded by

Copyright:

Available Formats

Lab 14

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lab 14

Uploaded by

Copyright:

Available Formats

Transaction Management

Database transactions are logical units of

Can have one of two outcomes:

All or nothing. All steps that are part of a

S= $300 S=$300-100 C=$800+100 S=$200

S= $300 S=$300-100 S=$300

Loss of T2’s update avoided by preventing

Loss of T2’s update avoided by preventing

Loss of T2’s update avoided by preventing

 Problem avoided by preventing T3 from reading

 Problem avoided by preventing T6 from reading balx

You might also like