Scribd
Upload
30 views5 pages

2018 - Midterm 1 Solution - Spring - LHR - ADB

Uploaded by

faizan majid
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
30 views5 pages

2018 - Midterm 1 Solution - Spring - LHR - ADB

Uploaded by

faizan majid
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

RollNo: ________________ Name: ___________________________

National University of Computer and Emerging Sciences, Lahore Campus


Course: Advanced Database Concepts Course Code: CS451
Program: BS(Computer Science) Semester: Spring 2018
Duration: 60 Minutes Total Marks: 34
Paper Date: 27-Feb-18 Weight 12.5%
Section: CS Page(s): 5
Exam: Midterm-I
Instruction/Notes: Scratch sheet can be used for rough work however, all the questions and steps are to be shown on
question paper. No extra/rough sheets should be submitted with question paper.
You will not get any credit if you do not show proper working, reasoning and steps as asked in
question statements.

Q1. (3 points) Given these transactions, find a cascade-free but not strict schedule, if
possible.
T1: r1(A), r1(B), w1(B), w1(A), c1.
T2: r2(B), w2(B), c2.
T3: w3(B), c3.

Ans:
r1(A), r1(B), r2(B), w3(B), w1(B), w2(B), w1(A), c1, c2, c3.

Q2. (6 points)
a) Describe how fuzzy checkpointing is used in ARIES.
b) What is the difference between the UNDO/REDO and the UNDO/NO-REDO algorithms for
recovery with immediate update?
c) Differentiate between non-repeatable read and phantom problems

Ans:
Get from text book.

School of Computer Science Page 1


RollNo: ________________ Name: ___________________________

Q3. (6 points) Consider the following classes of schedules: conflict-serializable, view-


serializable, recoverable, cascadeless, and strict. For each of the following schedules, state
which of the preceding classes it belongs to. If you cannot decide whether a schedule
belongs in a certain class based on the listed actions, explain briefly. The actions are listed
in the order they are scheduled. If a commit or abort is not shown, the schedule is
incomplete; assume that abort or commit must follow all the listed actions. Also draw
precedence graph for each schedule.
a) S1: r2(X), w3(X), c3, w1(Y), r2(Y), r2(Z), c2, r1(Z), c1.
b) S2: r2(X), w3(X), w1(Y), r2(Y), w2(Z)

Ans:
a) S1: r2(X), w3(X), c3, w1(Y), r2(Y), r2(Z), c2, r1(Z), c1.
S1: It is conflict-serializable, view serializable, not strict , not cascadeless, non-
recoverable as T2 read the dirty value of Y which is updated by T1 and T2 commit
before T1. Equivalent serial schedule is T1(Y)T2(X)T3.

b) S2: r2(X), w3(X), w1(Y), r2(Y), w2(Z)


S2: It is conflict-serializable, view serializable, not strict , not cascadeless. We
cannot decide whether it’s recoverable or not, since the abort/commit sequence
of these two transactions (i.e. T1 & T2 where T2 read dirty value of Y which is
updated by T1) are not specified. Equivalent serial schedule is T1T2T3.

School of Computer Science Page 2


RollNo: ________________ Name: ___________________________

Q4. (1+4+1+1= 7 points) Consider the execution shown in below Figure. Assume that the
Dirty Page Table and Transaction Table were empty before the start of the log.
a) What is the value of the LSN stored in the master log record?
b) What is done during Analysis? (Be precise about the points at which Analysis begins and
ends and show the contents of Dirty Page Table and Transaction Table constructed in this
phase.)
c) What is done during Redo? (Be precise about the points at which Redo begins and ends.)
d) What is done during Undo? (Be precise about the points at which Undo begins and ends.)

LSN LOG
1 begin_checkpoint
2 end_checkpoint
3 update: T1 writes C
4 update: T2 writes B
5 T1 commit
6 update: T3 writes A
7 update: T2 writes C
8 T2 commit
CRASH; RESTART

Ans:
a) LSN = 1
b) Analysis phase starts from LSN=2 record until it reaches the end i.e. LSN=8.
The end checkpoint contains the empty transaction table and dirty page table.
The analysis phase will further reconstruct these tables as follows:

Transaction Table (TT) Dirty Page Table (DPT)


TID Last_LSN Status Page_ID LSN
T1 35 inProcess commit C 3
T2 478 inProcess commit B 4
T3 6 inProcess A 6

c) REDO begins from LSN=3 (smallest LSN in DPT) and proceed with the REDO of
updates. The LSNs {3, 4, 6, 7} corresponding to the updates for pages C, B, A,
and C, respectively, are not less than the LSNs of pages in initial DPT. So those
data pages will be read again and all the updates reapplied from the LOG. At this
point REDO phase is finished and the UNDO phase starts.
d) UNDO is applied only to the active transaction T3. UNDO phase starts at LSN=6
(the last update of active transaction T3) and proceeds backward in the LOG. The
backward chain of updates for T3 (only LSN=6 record) is followed and undone.

School of Computer Science Page 3


RollNo: ________________ Name: ___________________________

Q5. (12 points) For the schedule S: r2(X), w3(X), c3, w1(Y), r2(Y), r2(Z), c2, r1(Z), c1. Show
that the schedule S will be accepted/rejected in exactly the order shown by the below
protocols. Provide proper reason and show your working.
a) Basic 2PL (add locks to the transactions)
b) Basic Timestamp Ordering (Assume T1 < T2 < T3)
c) Strict Timestamp Ordering (Assume T1 < T2 < T3)
d) Optimistic Concurrency Control

Ans:
a) Basic 2PL (Reject);
Here xl3(X) cannot acquire due to sl2(X).
1- ABORT T3 for Deadlock Avoidance using wait-die scheme
2- WAIT T3 for sl2(X) to release for Deadlock Avoidance using wound-wait
scheme
3- WAIT T3 for sl2(X) to release for Deadlock Detection (use wait-for-graph, in
case of deadlock)

b) Basic TO (ACCEPT);
- w3(X) allow as T3 is younger than RTS of X (i.e. TS(T3) > RTS(X) )
- r2(Y) allow as T2 is younger than WTS of Y (i.e. TS(T2) > WTS(Y) )

c) Strict TO (REJECT);
- w3(X) allow as T3 is younger than RTS of X (i.e. TS(T3) > RTS(X) )
- r2(Y) delay until c1/a1; as T2 is younger than WTS of Y (i.e. TS(T2) > WTS(Y) )

d) Optimistic(REJECT);
- T3 Forward Validation fails due to item X conflict with T2; so it may be delay
or abort T3/T2.
- But the validation of T1 & T2 is successful, if T3 abort.

School of Computer Science Page 4


RollNo: ________________ Name: ___________________________

School of Computer Science Page 5

You might also like