Unit4
1. Describe
the three main architectures of parallel DBMSs with neat diagram. Explain why shared
nothing architecture is more attractive than shared-memory and shared-disk approaches?
2. Define
the term scale–up and speed–up in parallel database systems. What can you say
about the speed-up and scale-up of the shared-nothing architecture?
3. Briefly
explain two kinds of interoperation parallelism that can be exploited within a
query. Explain the data partitioning and
parallelization of sequential operation evaluation in detail.
4. Discuss
joins in parallelizing individual operations.
Also explain improved parallel hash join.
5. What
are the functions that need to be provided by distributed databases in addition
to those of centralized DBMS ? What are
the different types of distributed database system?
6. Discuss
different types of data fragmentation in distributed database design. Distinguish
between horizontal and vertical fragmentations.
7. What
is the difference between synchronous and asynchronous replication with regard
to updating distributed data? Why has asynchronous replication gained in
popularity? Describe the voting and read- any write-all approaches to
synchronous replication. Also summarize the peer-to-peer and primary site
approaches to asynchronous replication.
8. Discuss
the query processing in distributed databases. Describe semijoins and
bloomjoins with respect to join of relations at different sites.
9.
Explain the distributed concurrency
control. Why distributed recovery is needed?
10.
Explain the need for a commit protocol
in a distributed DBMS. Explain 2 Phase commit Protocol in detail specifying the
list of messages initiated by 2PC.
No comments:
Post a Comment