Advance Database Management System

Q1.

Which of the following is not a parallel database architecture

Shared memory

Shared processor

Shared Disk

Shared nothing

Q2.

Supporting interquery parallelism is more complicated in______

Shared memory

shared disk

shared nothing

B and C

Q3.

Range partitioning works best for___

Large databases

small databases

Databases with less attributes

Databases with less range tuples

Q4.

What is the advantage of Intraquery parallelism?

Reduce network traffic

Reduce bottleneck

Speeding up long running queries

Scaling up the system

Q5.

__________designates one or more attributes from the given relations scheme as the partitioning attributes.

Round robin

Hash partitioning

Range partitioning

Vertical partitioning

Q6.

____________skew refers to the fact that there may be load imbalance in the partitioning, even when there is no attribute skew

Attribute value skew

Partition skew

Both

None of the above

Q7.

____________skew refers to the fact that some values appear in the partitioning attributes of many tuples

Attribute value skew

Partition skew

Both

None of the above

Q8.

Which are the two steps in range partitioning sort? i)Range partitioning sort ii) sorting each partition separately iii) Redistribute tuples iv) virtual partitioning

I,iii

I,iv,

I,ii

iii,iv

Q9.

Hash partitioning is a better choice than range than range partitioning when:

you don’t know beforehand how much data will map into a given range

sizes of range partitions would differ quite substantially.

Partition pruning and partition-wise joins on a partition-wise joins on a partitioning attribute are important

All of the above

Q10.

Round robin partitioning is ideally suited for which kind of applications?

Reading entire relation sequentially

reading some part of database

Random read operations

All of the above

Q11.

What are the issues motivating parallelism?

Increased availability

Distributed access to data

Analysis of distributed data

All of the above

Q12.

Which of the following is true in case of pipelining

They scale up well

It is possible to pipeline relational operators that donot produce O/P until I/P have been accessed.

Only marginal speedup is obtained for frequent cases in which one operator's execution cost is more than others'.

Pipeline chains attain sufficient length to provide high degree of parallelism

Q13.

In______________performance is sustained if the number of CPUs and disks are increased in proportion to the amount of data.

Linear scale up

Linear speed-up

Interference

None of the above

Q14.

The main advantage of shared memory and shared disk architecture is interference

TRUE

FALSE

cant say

Q15.

A__________ system seeks to improve performance through parallelization of various operations, such as loading data, building indexes and evaluating queries.

Distributed database

Parallel database

Client server database

standalone systems