Download to read offline
Uploaded byaineezafar2
PPT, PDF4 views
lecture 6 : introduction to concurency control.ppt
Concurrency control is a method used in databases to make sure that when many users access or change data at the same time, the data remains accurate, consistent, and free from conflicts.When multiple transactions run together (for example, many students updating their records at once), problems like lost updates, incorrect reads, or data corruption can occur. Concurrency control prevents these issues.
More Related Content
Similar to lecture 6 : introduction to concurency control.ppt
More from aineezafar2
lecture 6 : introduction to concurency control.ppt
- 1.
- 2.2Chapter 22 -Objectives Function and importance of transactions. Properties of transactions. Concurrency Control– Meaning of serializability.– How locking can ensure serializability.– Deadlock and how it can be resolved.– How timestamping can ensureserializability.– Optimistic concurrency control.
- 3.3Chapter 22 -Objectives Recovery Control– Some causes of database failure.– How to recover following databasefailure. Alternative models for long duration transactions.
- 4.4Transaction SupportTransactionAction, orseries of actions, carried out by user orapplication, which reads or updates contents ofdatabase. Logical unit of work on the database. Application program is series of transactions with non-database processing in between. Transforms database from one consistent state toanother, although consistency may be violated duringtransaction.
- 5.
- 6.6Transaction Support Canhave one of two outcomes:– Success - transaction commits and database reaches anew consistent state.– Failure - transaction aborts, and database must berestored 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 berestarted later.
- 7.7Properties of TransactionsFourbasic (ACID) properties of a transaction are:Atomicity ‘All or nothing’ property.Consistency Must transform database from one consistentstate to another.Isolation Partial effects of incomplete transactionsshould not be visible to other transactions.Durability Effects of a committed transaction arepermanent and must not be lost because of later failure.
- 8.8Concurrency ControlProcess ofmanaging simultaneous operations onthe database without having them interfere withone another. Prevents interference when two or more usersare accessing database simultaneously and atleast one is updating data. Although two transactions may be correct inthemselves, interleaving of operations mayproduce an incorrect result.
- 9.9Need for ConcurrencyControl Three examples of potential problems caused byconcurrency:– Lost update problem.– Uncommitted dependency problem.– Inconsistent analysis problem.
- 10.10Lost Update ProblemSuccessfully completed update is overridden byanother user. T1 withdrawing £10 from an account with balx,initially £100. T2 depositing £100 into same account. Serially, final balance would be £190.
- 11.11Lost Update ProblemLoss of T2’s update avoided by preventing T1from reading balx until after update.
- 12.12Uncommitted Dependency ProblemOccurs when one transaction can seeintermediate results of another transactionbefore it has committed. T4 updates balx to £200 but it aborts, so balxshould be back at original value of £100. T3 has read new value of balx (£200) and usesvalue as basis of £10 reduction, giving a newbalance of £190, instead of £90.
- 13.13Uncommitted Dependency ProblemProblem avoided by preventing T3 fromreading balx until after T4 commits or aborts.
- 14.14Inconsistent Analysis ProblemOccurs when transaction reads several values butsecond transaction updates some of them duringexecution of first. Sometimes referred to as dirty read or unrepeatableread. T6 is totaling balances of account x (£100), accounty (£50), and account z (£25). Meantime, T5 has transferred £10 from balx tobalz, so T6 now has wrong result (£10 too high).
- 15.15Inconsistent Analysis ProblemProblem avoided by preventing T6 from reading balx and balzuntil after T5 completed updates.
- 16.16Serializability Objective ofa concurrency control protocol is toschedule transactions in such a way as to avoidany interference. Could run transactions serially, but this limitsdegree of concurrency or parallelism in system. Serializability identifies those executions oftransactions guaranteed to ensure consistency.
- 17.17SerializabilityScheduleSequence of reads/writesby set of concurrenttransactions.Serial ScheduleSchedule where operations of each transactionare executed consecutively without anyinterleaved operations from other transactions. No guarantee that results of all serial executionsof a given set of transactions will be identical.
- 18.18Nonserial Schedule Schedulewhere operations from set of concurrenttransactions are interleaved. Objective of serializability is to find nonserialschedules that allow transactions to executeconcurrently without interfering with one another. In other words, want to find nonserial schedulesthat are equivalent to some serial schedule. Such aschedule is called serializable.
- 19.19Serializability In serializability,ordering of read/writes isimportant:(a) If two transactions only read a data item, theydo not conflict and order is not important.(b) If two transactions either read or writecompletely separate data items, they do notconflict and order is not important.(c) If one transaction writes a data item andanother reads or writes same data item, orderof execution is important.
- 20.20Example of ConflictSerializability
- 21.21Serializability Conflict serializableschedule orders anyconflicting operations in same way as some serialexecution. Under constrained write rule (transaction updatesdata item based on its old value, which is firstread), use precedence graph to test forserializability.
- 22.22Precedence Graph Create:–node for each transaction;– a directed edge Ti Tj, if Tj reads the value ofan item written by TI;– a directed edge Ti Tj, if Tj writes a value intoan item after it has been read by Ti. If precedence graph contains cycle schedule isnot conflict serializable.
- 23.23Example - Non-conflictserializable schedule T9 is transferring £100 from one account withbalance balx to another account with balancebaly. T10 is increasing balance of these two accounts by10%. Precedence graph has a cycle and so is notserializable.
- 24.24Example - Non-conflictserializable schedule
- 25.25View Serializability Offersless stringent definition of schedule equivalencethan conflict serializability. Two schedules S1 and S2 are view equivalent if:– For each data item x, if Ti reads initial value of x in S1, Timust also read initial value of x in S2.– For each read on x by Ti in S1, if value read by x iswritten by Tj, Ti must also read value of x produced by Tjin S2.– For each data item x, if last write on x performed by Ti inS1, same transaction must perform final write on x in S2.
- 26.26View Serializability Scheduleis view serializable if it is viewequivalent to a serial schedule. Every conflict serializable schedule is viewserializable, although converse is not true. It can be shown that any view serializableschedule that is not conflict serializable containsone or more blind writes. In general, testing whether schedule isserializable is NP-complete.
- 27.27Example - ViewSerializable schedule
- 28.28Recoverability Serializability identifiesschedules that maintaindatabase consistency, assuming no transactionfails. Could also examine recoverability oftransactions within schedule. If transaction fails, atomicity requires effects oftransaction to be undone. Durability states that once transaction commits,its changes cannot be undone (without runninganother, compensating, transaction).
- 29.29Recoverable ScheduleA schedulewhere, for each pair of transactionsTi and Tj, if Tj reads a data item previouslywritten by Ti, then the commit operation of Tiprecedes the commit operation of Tj.