Code No: 45027
Uploaded by
SRINIVASA RAO GANTACode No: 45027
Uploaded by
SRINIVASA RAO GANTACode No: 45027 R07 Set No - 1
III B.Tech I Semester Regular Examinations,Nov/Dec 2009
DISTRIBUTED DATABASES
Information Technology
Time: 3 hours Max Marks: 80
Answer any FIVE Questions
All Questions carry equal marks
?????
1. Discuss general queries and optimization methods for general queries. [16]
2. Explain multi database concurrency control algorithms in detail. [16]
3. (a) What is check point? Explain its roles in recovery procedures. [8]
(b) Discuss remote recovery information problem in detail. [8]
4. (a) Describe distributed garbage collection. [8]
(b) Give the architecture of page servers used for distributed system. [8]
5. (a) What is a false deadlock? Discuss the approaches to the problem of false
deadlock. [8]
(b) Explain the requirements of timestamps in detail. [8]
6. Explain the trade offs of write-locks-all, majority locking and primary copy locking
approaches in a redundant databases. [16]
7. Assume the global and fragmentation schema
PATIENT (PNUM, NAME, DEPT, TREAT, DNUM)
CARE (PNUM, DRUG, QUAN)
PATIENT1=SLDEP T =“SU RGERY ”&T REAT =“IN T EN SIV E” PATIENT
PATIENT2=SLDEP T =“SU RGERY ”&T REAT 6=“IN T EN SIV E” PATIENT
PATIENT3=SLDEP T 6=“SU RGERY ” PATIENT
CARE1= CARE SJP N U M =P N U M PATIENT1
CARE2= CARE SJP N U M =P N U M PATIENT2
CARE3= CARE SJP N U M =P N U M PATIENT3
Translate the following global queries into fragment queries and simplify them. Use
the algebra of qualified relations.
(a) PJT REAT SLDRU G=“N IM SU LIDE” CARE NJN PATIENT). [8]
(b) GBAV G(QU AN ) SLDRU G=“ASP RIN ” (CARE NJN SLT REAT =“IN T EN SIV E” PATIENT).
[8]
8. Consider the following global, fragmentation and allocation schemata :
Global schema: SUPPLIER (SNUM, NAME, CITY)
Fragmentation schema: SUPPLIER1 = SLCIT Y =“BAN GALORE” SUPPLIER
SUPPLIER2 = SLCIT Y =“HY DERABAD” SUPPLIER
Allocation schema: SUPPLIER1 at site 1
SUPPLIER2 at site 1, site 2
1
Code No: 45027 R07 Set No - 1
Write an application that moves a SUPPLIER having ID 522 from Bangalore to
Hyderabad at levels 1, 2 & 3 transparency. [16]
?????
2
Code No: 45027 R07 Set No - 2
III B.Tech I Semester Regular Examinations,Nov/Dec 2009
DISTRIBUTED DATABASES
Information Technology
Time: 3 hours Max Marks: 80
Answer any FIVE Questions
All Questions carry equal marks
?????
1. (a) Explain search space and transformation rules for query processing. . [8]
(b) Discuss in detail object identifier management. [8]
2. (a) Explain termination protocols for 3-phase commitment. [8]
(b) List and explain the rules of quorum based protocol. [8]
3. Give and explain reference architecture for distributed databases. [16]
4. (a) Write a note on N-ary integration methods. [5]
(b) Give the structure of a distributed Multi-DBMS. [5]
(c) Discuss the object models characteristics that are important in addressing
interoperability. [6]
5. (a) List and explain the rules which define the result of applying the operations
of relational algebra to qualified relations. [10]
(b) Explain operator graph of a semi join operation. [6]
6. Explain use of joins as a query processing tactic. [16]
7. (a) Explain transformation of centralized 2 PC protocol into a distributed commit
protocol. [8]
(b) Discuss clarification of reliability algorithms. [4]
(c) Give a note on increasing efficiency of basic 2 PC protocol. [4]
8. What is a correct distributed concurrency control mechanism? List some correct
distributed concurrency control mechanisms. Prove that timestamp based protocol
is correct. [16]
?????
3
Code No: 45027 R07 Set No - 3
III B.Tech I Semester Regular Examinations,Nov/Dec 2009
DISTRIBUTED DATABASES
Information Technology
Time: 3 hours Max Marks: 80
Answer any FIVE Questions
All Questions carry equal marks
?????
1. (a) Give a note on user identification in distributed database administration. [5]
(b) Write about Byzantine agreement. [5]
(c) Explain primary copy locking approach in a redundant databases. [6]
2. Explain briefly a framework for query optimization with suitable illustrations. [16]
3. Explain distribution transparency for update applications. [16]
4. (a) Explain the database integration process. [8]
(b) Discuss the database interoperability in the COM/OLE environment. [8]
5. (a) Explain the basic storage models for object clustering. [8]
(b) Give and explain the architecture of object servers. Discuss its advantage and
limitations. [8]
6. (a) Explain in detail recovery of distribution transactions in distribution databases.
[10]
(b) “Isolation is a less desirable property of a transaction”. Support this state-
ment. [6]
7. Compute the following expression, using the algebra of qualified relations. Explain
in detail the process of computation.
PJR3.T SLb ((([R1: c] CP [R2: NOT b]) UN ([R1: c] JNF [R2: NOT b])) DF ([R3:
d] JNF [S2: a])). [16]
8. (a) Show that “occurred before” relationship in a two site system is partial order-
ing. [8]
(b) Explain how to derive a potential global deadlock cycle from a local wait-for-
graph. [8]
?????
4
Code No: 45027 R07 Set No - 4
III B.Tech I Semester Regular Examinations,Nov/Dec 2009
DISTRIBUTED DATABASES
Information Technology
Time: 3 hours Max Marks: 80
Answer any FIVE Questions
All Questions carry equal marks
?????
1. (a) Explain validation using only transaction timestamps. [8]
(b) Give a detail note on basic timestamp mechanism. [8]
2. (a) Discuss simplification of parametric queries. [8]
(b) Explain algebra of qualified relations. [8]
3. (a) Describe the model of a distributed transaction. [8]
(b) Explain the basic of 2-phase commitment protocol. [8]
4. List and explain the problems which arise in designing a reliable distributed database
system. [16]
5. (a) List and explain the important requirements for transaction management in
object DBMS. [8]
(b) Explain management of aggregation graph in detail. [8]
6. (a) Write a note on resolving structural conflicts. [8]
(b) Explain the structure of distributed multi-DBMS. [8]
7. Consider the following global, fragmentation and allocation schemata :
Global schema: CUSTOMER (CNO, CNAME, CCITY)
Fragmentation schema: CUSTOMER1 = SLCIT Y =“DELHI” CUSTOMER
CUSTOMER2 = SLCIT Y =“M U M BAI” CUSTOMER
CUSTOMER3 = SLCIT Y =“RAN CHI” CUSTOMER
Allocation schema: CUSTOMER1 at sites 1, 3
CUSTOMER2 at sites 2, 4
CUSTOMER2 at sites 1, 5
Write an application that requires customer number from the terminal and output
the name and city of the customer at level 3 transparency. [16]
8. (a) Discuss about integration of scheduler in AHY approach. [8]
(b) Write a note on effect of commuting joins & unions. [8]
?????