7/31/2019 UNIT 1 (ADBMS)

1/35

Unit 1

Transactions

1

7/31/2019 UNIT 1 (ADBMS)

2/35

Transactions

Transaction Concept

Transaction State

Concurrent Execution of Transactions

Schedules

Serializability

Testing for Serializability Recoverability

2

7/31/2019 UNIT 1 (ADBMS)

3/35

Transaction Concept

Transaction is collection of operations that form a single logicalunit of program execution.

A transactionis a unitof program execution that accesses andpossibly updates various data items.

E.g. transaction to transfer Rs.500 from account A to account B:

1. read(A)

2. A := A 500

3. write(A)

4. read(B)

5. B:= B +500

6. write(B)

3

7/31/2019 UNIT 1 (ADBMS)

4/35

Features of Transaction

1. It is initiated by a user program written in DML orprogramming language ( SQL, C, C++).

2. Transaction specifies all operations executedbetween the begin and end transaction.

3. It uses two operations: read and write.

4

7/31/2019 UNIT 1 (ADBMS)

5/35

Operations of Transaction

Read(X) : It transfers the data item Xfrom the database to a local buffer inmain memory.

Write(X): It transfers the data item Xfrom the local buffer back to thedatabase.

5

7/31/2019 UNIT 1 (ADBMS)

6/35

ACID Properties

Atomicity. Either all operations of the transaction are properly reflectedin the database or none are.

Consistency. Execution of a transaction in isolation preserves the

consistency of the database. Isolation. Although multiple transactions may execute concurrently,

each transaction must be unaware of other concurrently executingtransactions. Intermediate transaction results must be hidden from otherconcurrently executed transactions.

That is, for every pair of transactions Tiand Tj, it appears to Tithateither Tj, finished execution before Tistarted, or Tjstarted executionafter Tifinished.

Durability. After a transaction completes successfully, the changes ithas made to the database persist, even if there are system failures.

A transaction is a unit of program execution that accesses and possibly

updates various data items.To preserve the integrity of data the databasesystem must ensure:

6

P i f T i i h E l f F d

7/31/2019 UNIT 1 (ADBMS)

7/35

Properties of Transaction with Example of FundTransfer

Transaction to transfer Rs.500 from account A to account B:

1. read(A)2. A := A 500

3. write(A)

4. read(B)

5. B:= B +500

6. write(B)

Atomicity requirement

if the transaction fails after step 3 and before step 6, money will be lost

leading to an inconsistent database state

Failure could be due to software or hardware

the system should ensure that updates of a partially executed transaction

are not reflected in the database Durability requirement once the user has been notified that the transaction

has completed (i.e., the transfer of the Rs. 500 has taken place), the updatesto the database by the transaction must persist even if there are software orhardware failures.

7

7/31/2019 UNIT 1 (ADBMS)

8/35

Example of Fund Transfer (Cont.)

Transaction to transfer Rs.500 from account A to account B:1. read(A)

2. A := A 500

3. write(A)

4. read(B)

5. B:= B +500

6. write(B)

Consistency requirement in above example:

the sum of A and B is unchanged by the execution of the transaction

A transaction must see a consistent database.

During transaction execution the database may be temporarily inconsistent.

When the transaction completes successfully the database must be consistent

Erroneous transaction logic can lead to inconsistency

8

7/31/2019 UNIT 1 (ADBMS)

9/35

Example of Fund Transfer (Cont.)

Isolation requirement if between steps 3 and 6, another

transaction T2 is allowed to access the partially updated database, itwill see an inconsistent database (the sum A + Bwill be less than itshould be).

T1 T2

1. read(A)

2. A := A 500

3. write(A) read(A), read(B), print(A+B)

4. read(B)

5. B:= B +500

6. write(B

Isolation can be ensured trivially by running transactions serially

that is, one after the other.

However, executing multiple transactions concurrently has significantbenefits, as we will see later.

9

7/31/2019 UNIT 1 (ADBMS)

10/35

Transaction State

Activethe initial state; the transaction stays in this state while it is

executing Partially committedafter the final statement has been executed.

Failed -- after the discovery that normal execution can no longerproceed.

Aborted after the transaction has been rolled back and the

database restored to its state prior to the start of the transaction.Two options after it has been aborted:

restart the transaction

can be done only if no internal logical error

kill the transaction

Committed after successful completion.

10

7/31/2019 UNIT 1 (ADBMS)

11/35

Transaction State (Cont.)

11

I l t ti f At i it d

7/31/2019 UNIT 1 (ADBMS)

12/35

Implementation of Atomicity andDurability

The recovery-management component of a database systemimplements the support for atomicity and durability.

E.g. the shadow-databasescheme:

all updates are made on a shadow copyof the database

db_pointer is made to point to the updated shadow copy after

the transaction reaches partial commit and

all updated pages have been flushed to disk.

12

7/31/2019 UNIT 1 (ADBMS)

13/35

Implementation of Atomicity and Durability(Cont.)

db_pointer always points to the current consistent copy of the database. In case transaction fails, old consistent copy pointed to by db_pointer

can be used, and the shadow copy can be deleted.

The shadow-database scheme:

Assumes that only one transaction is active at a time.

Assumes disks do not fail

Useful for text editors, but

extremely inefficient for large databases (why?)

Variant called shadow paging reduces copying of data, but isstill not practical for large databases

Does not handle concurrent transactions

13

7/31/2019 UNIT 1 (ADBMS)

14/35

Concurrent Execution of Transactions

Multiple transactions are allowed to run concurrently in the system.

Advantages are: increased processor and disk utilization, leading to better

transaction throughput

E.g. one transaction can be using the CPU while another isreading from or writing to the disk

reduced average response time for transactions: shorttransactions need not wait behind long ones.

Concurrency control schemes mechanisms to achieve isolation

that is, to control the interaction among the concurrenttransactions in order to prevent them from destroying the

consistency of the database Details will be studied later in unit 2.

14

7/31/2019 UNIT 1 (ADBMS)

15/35

Schedules

Schedule a sequences of instructions that specify the chronological

order in which instructions of concurrent transactions are executed.

a schedule for a set of transactions must consist of all instructionsof those transactions

must preserve the order in which the instructions appear in each

individual transaction.

A transaction that successfully completes its execution will have acommit instructions as the last statement

by default transaction assumed to execute commit instruction as its

last step

A transaction that fails to successfully complete its execution will havean abort instruction as the last statement

15

7/31/2019 UNIT 1 (ADBMS)

16/35

Schedule 1

Let T1 transfer $50 from A to B, and T2 transfer 10% of the

balance from A to B. A serial schedule in which T1 is followed by T2:

16

7/31/2019 UNIT 1 (ADBMS)

17/35

Schedule 2

A serial schedule where T2 is followed by T1

17

7/31/2019 UNIT 1 (ADBMS)

18/35

Schedule 3

Let T1 and T2 be the transactions defined previously. The

following schedule is not a serial schedule, but it is equivalentto Schedule 1.

In Schedules 1, 2 and 3, the sum A + B is preserved.

18

7/31/2019 UNIT 1 (ADBMS)

19/35

Schedule 4

The following concurrent schedule does not preserve the

value of (A + B).

19

7/31/2019 UNIT 1 (ADBMS)

20/35

Serializability

Basic Assumption Each transaction preserves database

consistency. Thus serial execution of a set of transactions preserves database

consistency.

A (possibly concurrent) schedule is serializable if it is equivalent to aserial schedule. Different forms of schedule equivalence give rise to

the notions of:1. conflict serializability

2. view serializability

Simplified view of transactions

We ignore operations other than read and write instructions

We assume that transactions may perform arbitrary computationson data in local buffers in between reads and writes.

Our simplified schedules consist of only read and writeinstructions.

20

7/31/2019 UNIT 1 (ADBMS)

21/35

Conflicting Instructions

Instructions liand ljof transactions Tiand Tj

respectively, conflict if and only if there exists someitem Qaccessed by both liand lj, and at least one ofthese instructions write(Q).

1. li= read(Q), lj=read(Q). liand ljdont conflict.

2. li= read(Q), lj=write(Q). They conflict.3. li= write(Q), lj=read(Q). They conflict4. li= write(Q), lj=write(Q). They conflict

21

7/31/2019 UNIT 1 (ADBMS)

22/35

Conflict Serializability

If a schedule Scan be transformed into aschedule Sby a series of swaps of non-conflicting instructions, we say that Sand Sare conflict equivalent.

We say that a schedule Sis conflictserializable if it is conflict equivalent to a

serial schedule

22

7/31/2019 UNIT 1 (ADBMS)

23/35

Conflict Serializability (Cont.)

Schedule 3 can be transformed into Schedule 6, a serial

schedule where T2 follows T1, by series of swaps of non-conflicting instructions.

Therefore Schedule 3 is conflict serializable.

Schedule 3Schedule 6

23

7/31/2019 UNIT 1 (ADBMS)

24/35

Conflict Serializability (Cont.)

Example of a schedule that is not conflict serializable:

We are unable to swap instructions in the above schedule to obtaineither the serial schedule < T3, T4 >, or the serial schedule < T4, T3 >.

24

7/31/2019 UNIT 1 (ADBMS)

25/35

View Serializability

Let Sand Sbe two schedules with the same set of transactions. S

and Sare view equivalentif the following three conditions are met,for each data item Q,

1. If in schedule S, transaction Tireads the initial value of Q, then inschedule Salso transaction Ti must read the initial value of Q.

2. If in schedule S transaction Tiexecutes read(Q), and that value

was produced by transaction Tj (if any), then in scheduleS

alsotransaction Timust read the value of Qthat was produced by thesame write(Q) operation of transaction Tj.

3. The transaction (if any) that performs the final write(Q) operationin schedule Smust also perform the finalwrite(Q) operation inschedule S.

As can be seen, view equivalence is also based purely on reads andwrites alone.

25

7/31/2019 UNIT 1 (ADBMS)

26/35

View Serializability (Cont.)

A schedule Sis view serializableif it is view equivalent to a serial

schedule. Every conflict serializable schedule is also view serializable.

Below is a schedule which is view-serializable but notconflictserializable.

What serial schedule is above equivalent to?

Every view serializable schedule that is not conflict serializable hasblind writes.

26

7/31/2019 UNIT 1 (ADBMS)

27/35

Other Notions of Serializability

The schedule below produces same outcome as the serial

schedule < T1,T5 >, yet is not conflict equivalent or viewequivalent to it.

Determining such equivalence requires analysis of operationsother than read and write.

27

7/31/2019 UNIT 1 (ADBMS)

28/35

Testing for Serializability

Consider some schedule of a set of transactions T1, T2, ..., Tn

Precedence graph a direct graph where the vertices arethe transactions (names).

We draw an arc from Tito Tjif the two transaction conflict,and Tiaccessed the data item on which the conflict aroseearlier.

We may label the arc by the item that was accessed. Example 1

x

y

28

7/31/2019 UNIT 1 (ADBMS)

29/35

Example Schedule (Schedule A) + Precedence Graph

T1 T2 T3 T4 T5

read(X)

read(Y)

read(Z)

read(V)

read(W)

read(W)

read(Y)

write(Y)

write(Z)

read(U)

read(Y)

write(Y)

read(Z)

write(Z)

read(U)

write(U)

T3T4

T1 T2

T5

29

7/31/2019 UNIT 1 (ADBMS)

30/35

Test for Conflict Serializability

A schedule is conflict serializable if and only

if its precedence graph is acyclic. Cycle-detection algorithms exist which take

order n2 time, where nis the number ofvertices in the graph.

(Better algorithms take order n+ ewhere eis the number of edges.)

If precedence graph is acyclic, theserializability order can be obtained by atopological sortingof the graph.

This is a linear order consistent with thepartial order of the graph.

For example, a serializability order forSchedule A would beT5T1T3T2T4

Are there others?

30

7/31/2019 UNIT 1 (ADBMS)

31/35

Testing for View Serializability

31

7/31/2019 UNIT 1 (ADBMS)

32/35

Test for View Serializability

The precedence graph test for conflict serializability cannot be used

directly to test for view serializability. Extension to test for view serializability has cost exponential in the

size of the precedence graph.

The problem of checking if a schedule is view serializable falls in theclass of NP-complete problems.

Thus existence of an efficient algorithm is extremelyunlikely. However practical algorithms that just check some sufficient

conditions for view serializability can still be used.

32

Recoverability

7/31/2019 UNIT 1 (ADBMS)

33/35

Recoverability

Recoverableschedule if a transaction Tjreads a data itempreviously written by a transaction Ti, then the commit operation of Tiappears before the commit operation of Tj.

The following schedule (Schedule 11) is not recoverable if T9commitsimmediately after the read

If T8should abort, T9 would have read (and possibly shown to the user)an inconsistent database state. Hence, database must ensure thatschedules are recoverable.

Need to address the effect of transaction failures on concurrently

running transactions.

33

7/31/2019 UNIT 1 (ADBMS)

34/35

Cascading Rollbacks

Cascading rollback a single transaction failure leads to a

series of transaction rollbacks. Consider the following schedulewhere none of the transactions has yet committed (so theschedule is recoverable)

If T10 fails, T11 and T12 must also be rolled back. Can lead to the undoing of a significant amount of work

34

7/31/2019 UNIT 1 (ADBMS)

35/35

Cascadeless Schedules

Cascadelessschedules cascading rollbacks cannot occur; for

each pair of transactions Tiand Tjsuch that Tj reads a data itempreviously written by Ti, the commit operation of Ti appears before theread operation of Tj.

Every cascadeless schedule is also recoverable

It is desirable to restrict the schedules to those that are cascadeless

35