cs4432: database systems ii

cs4432 concurrency control 1

CS4432: Database Systems II

Concurrency Control with Recovery


Example: Ti To

Wi(A) rj(A) Commit To

Abort Ti

Concurrency control & recovery

……

… ……

…Cascading rollback ( But already committed!)


Note :

Schedule is conflict serializable, But schedule is not recoverable.


• Need to make “final’ decision for each transaction:

– commit decision - system guarantees transaction will or has completed, no matter what (Ci)

– abort decision - system guarantees transaction will or has been rolled back

(has no effect) (Ai)


...

...

...

...

Back to example:

Ti Tj

Wi(A)rj(A)

Cj can we commit

here?


Definition

Ti read from Tj in S (Tj S Tj) if

(1)wj(A) <S ri(A)

(2)aj <S ri(A) (< : does not precede)

(3) If wj(A) <S wk(A) <S ri(A) then

ak <S ri(A)


Definition

Intuition : A schedule S is recoverable,

if transactions only commit after all transactions they read from have already been committed first.

Formal : Schedule S is recoverable if whenever Tj S Ti and j i and Ci S

then Cj <S Ci


How to achieve recoverable schedules?


Strict 2PL : With 2PL, hold write locks until commit

Ti Tj

Wi(A)

Ciui(A)

rj(A)

...

...

...

...

...

...

...


• S is recoverable if each transaction commits only after all transactions from which it read have committed.

• S avoids cascading rollback if each transaction may read only those values written by committed transactions.


• S is strict if each transaction may read and write only items previously written by committed transactions.

Avoids cascading rollback

RC

ACR

STSERIAL

14

Examples

• Recoverable:– w1(A) w1(B) w2(A) r2(B) c1 c2

• Avoids Cascading Rollback:– w1(A) w1(B) w2(A) c1 r2(B) c2

• Strict:– w1(A) w1(B) c1 w2(A) r2(B) c2

Assumes w2(A) is done

without reading

CS4432 transaction management 15

Schedule with Deadlock

T1 T2l1(A); Read(A) l2(B);Read(B)A A+100;Write(A) B Bx2;Write(B)

ll11(B)(B) l l22(A)(A)delayeddelayed


Deadlocks

• Detection– Wait-for graph

• Prevention– Resource ordering– Timeout– Wait-die– Wound-wait


Deadlock Detection

• Build Wait-For graph• Use lock table structures• Build incrementally or periodically• When cycle found, rollback victim

T1

T3

T2

T6

T5

T4T7


Resource Ordering

• Order all elements A1, A2, …, An

• A transaction T can lock Ai after Aj only if i > j

Problem : Ordered lock requests not realistic in most cases


Timeout

• If transaction waits more than L sec., roll it back!

• Simple scheme• Hard to select L


Wait-die

• Transactions given a timestamp when they arrive …. ts(Ti)

• Ti can only wait for Tj if ts(Ti)< ts(Tj) ...else die


T1

(ts =10)

T2

(ts =20)

T3

(ts =25)

wait

wait

Example:

wait?


Wound-wait

• Transactions given a timestamp when they arrive … ts(Ti)

• Ti wounds Tj if ts(Ti)< ts(Tj) else Ti waits

“Wound”: Tj rolls back and gives lock to Ti


Summary

• Cascading rollbackRecoverable schedule

• Deadlock– Prevention– Detection

Overall Summary


25

Recovery: DB Dump + Log

backup

database

active

databaselog

• If active database is lost,

– restore active database from backup

– bring up-to-date using redo entries in log

26

When can log be discarded?

check-

point

db

dump

last

needed

undo

not needed for

media recovery

not needed for undo

after system failure

not needed for

redo after system failure

log

time

27

Multiple Mode Locks: Shared locksSo far:S = ...l1(A) r1(A) u1(A) … l2(A) r2(A) u2(A) …

Do not conflict

Instead:S=... ls1(A) r1(A) ls2(A) r2(A) …. us1(A)

us2(A)

Also, we have exclusive (X)/write locks

28

Ti

Read(A),Write(B)

l(A),Read(A),l(B),Write(B)…

Read(A),Write(B)

Scheduler, part I

Scheduler, part II

DB

locktable

Locks in Practice

29

Lock table Conceptually

A

BC

...

Lock info for B

Lock info for C

If null, object is unlocked

Every

poss

ible

obje

ct

30

But use hash table:

A

If object not found in hash table, it is unlocked

Lock info for AA

......

H

31

Lock info for A - example

tran mode wait? Nxt T_link

Object:AGroup mode:UWaiting:yesList:

T1 S no

T2 U no

T3 XX yes

To other T3

records

32

What are the objects we lock?

?

Relation A

Relation B

...

Tuple A

Tuple BTuple C

...

Disk block

A

Disk block

B

...

DB DB DB

33

• Locking works in any case, but should we choose small or large objects?

• If we lock large objects (e.g., Relations)– Need few locks– Low concurrency

• If we lock small objects (e.g., tuples,fields)– Need more locks– More concurrency

34

We can have it both ways!!

Ask any janitor to give you the solution...

hall

Stall 1 Stall 2 Stall 3 Stall 4

restroom

35

Example

R1

t1t2 t3

t4

T1(IS)

T1(S)

, T2(S)

36

Example

R1

t1t2 t3

t4

T1(IS)

T1(S)

, T2(IX)

T2(IX)

37

Multiple granularity

Comp Requestor

IS IX S SIX X IS

Holder IX S

SIX

X

38

Multiple granularity

Comp Requestor

IS IX S SIX X IS

Holder IX S

SIX

X

T T T T F

F

F

F

FFFFF

FFFT

FTFT

FFTT

39

Inserting Records: PhantomsPhantoms

Example: relation R (E#,name,…)constraint: E# is keyuse tuple locking

R E# Name ….o1 55 Smitho2 75 Jones

40

T1: Insert <04,Kerry,…> into RT2: Insert <04,Bush,…> into R

T1 T2

S1(o1) S2(o1)S1(o2) S2(o2)Check Constraint Check Constraint

Insert o3[04,Kerry,..] Insert

o4[04,Bush,..]

... ...

41

Solution

• Use multiple granularity tree• Before insert of node Q, lock parent(Q) in X mode R1

t1t2 t3

42

Back to exampleT1: Insert<04,Kerry> T2: Insert<04,Bush>

T1 T2

X1(R)

Check constraintInsert<04,Kerry>U(R)

X2(R)Check constraintOops! e# = 04 already in

R!

XX22(R(R))

delayed

43

Instead of using R, can use index on R:

Example: R

Index0<E#<100

Index100<E#<200

E#=2 E#=5 E#=107 E#=109...

...

...

cs4432: database systems ii

Documents