Distributed Database Question Bank with Answers 07

Question:

Explain the differences between distributed, parallel, and federated databases

Answer:

A distributed DBMS (DDBMS) is a DBMS where different parts of the database is located at different geographic locations (i.e. in several different computers or clusters) but where each part of the database is managed by local instances of the same DDBMS.

A parallel DBMS (PDBMS) is a single DBMS running on a single cluster where the PDBMS engine automatically decides how data is distributed and queries parallelized.

A federated DBMS (FDBMS) is a middleware DBMS that combines data from several different autonomous underlying DBMSs. SQL queries to the FDBMS are more or less automatically translated into queries to the underlying DBMSs.

**********************

Related Questions:

• Go to Distributed Database Quiz Questions Home page
• Go to Distributed Database Home page

• What is a distributed database? What is it not? If all sites use a DBMS from the same vendor, what do we call it?

• List two ways to measure the time improvement when processing a query using parallel architecture? Why is linearity important

• Describe the main advantage of pipelined parallelism and some of its drawbacks.

• When and why is replication favorable in a distributed database?

• When and why is fragmentation favorable in a distributed database?

• Compare and analyze the relative merits of centralized and hierarchical deadlock detection approaches in a distributed DBMS.

• Explain the differences between distributed, parallel, and federated databases

• What different kinds of transparencies are handled by distributed, parallel, and federated databases?

• List few situations in which adaptive query processing is beneficial.

• How is autonomy handled by distributed, parallel, and federated databases?

 

Differentiate between distributed, parallel and federated databases

What are the major differences between distributed, parallel and federated databases

List the differences in distributed database

Distributed Database Question Bank with Answers 08

Question:

What different kinds of transparencies are handled by distributed, parallel, and federated databases?  

Answer:

Transparencies in distributed database:

• In a distributed database the location of data is designed manually by the DBA. The distribution is specified using conditions to the DDBMS that declaratively specify how to redistribute data in tables. Thus distributed databases do not have schema transparency, since the distributed schema is designed manually.

• In a distributed database queries and updates are specified without need to know the names of the tables in the individual parts of the database. This is called naming transparency.

• In a distributed database, the user need not know how to split a query or an update into different subqueries to the different parts of the database. This is called query and update transparency.

• In summary, distributed databases have query, update, and naming transparency, but no schema transparency. 

Transparencies in parallel database:

• Parallel databases have schema, query, update, and naming transparency. Parallel databases run on clusters, which enables the parallel DMBS to automatically place table fragments. 

Transparencies in federated database:

• Federated databases have query and naming transparency, but limited updating and no schema transparency. Updating the integrated underlying local databases has to be done locally. In the federated database only tables stored in the global conceptual schema can be updated. All participating database have separate schemas.

**********************

Related Questions:

• Go to Distributed Database Quiz Questions Home page
• Go to Distributed Database Home page

• What is a distributed database? What is it not? If all sites use a DBMS from the same vendor, what do we call it?

• List two ways to measure the time improvement when processing a query using parallel architecture? Why is linearity important

• Describe the main advantage of pipelined parallelism and some of its drawbacks.

• When and why is replication favorable in a distributed database?

• When and why is fragmentation favorable in a distributed database?

• Compare and analyze the relative merits of centralized and hierarchical deadlock detection approaches in a distributed DBMS.

• Explain the differences between distributed, parallel, and federated databases

• List few situations in which adaptive query processing is beneficial.

• How is autonomy handled by distributed, parallel, and federated databases?

 

List the transparencies shown by distributed database

List the transparencies supported by parallel database

List the transparencies shown by federated database

what are the transparencies supported and declined by distributed, parallel and federated database

Distributed Database Question Bank with Answers 09

Question:

List three situations in which adaptive query processing is beneficial.

Answer:

What is adaptive query processing?

Adaptive Query Processing (AQP) analyzes actual query run time statistics and uses that information for subsequent optimizations. It focuses on using runtime feed-back to modify query processing in a way that provides better response time or more efficient CPU utilization.

AQP addresses the problems of missing statistics, unexpected correlations, unpredictable costs, and dynamic data by using feedback to tune execution. 

Situations in which AQP is beneficial:

• In situations where statistics are inaccurate or unavailable – e.g., query processing over a remotely stored web site.

• In situations where statistics change over the lifetime of the query – either because the query is long running (e.g., continuous) or stats are volatile.

• In situations with variable or unpredictable resources e.g., query processing on a cluster of shared machines

**********************

Related Questions:

• Go to Distributed Database Quiz Questions Home page
• Go to Distributed Database Home page

• What is a distributed database? What is it not? If all sites use a DBMS from the same vendor, what do we call it?

• List two ways to measure the time improvement when processing a query using parallel architecture? Why is linearity important

• Describe the main advantage of pipelined parallelism and some of its drawbacks.

• When and why is replication favorable in a distributed database?

• When and why is fragmentation favorable in a distributed database?

• Compare and analyze the relative merits of centralized and hierarchical deadlock detection approaches in a distributed DBMS.

• Explain the differences between distributed, parallel, and federated databases

• What different kinds of transparencies are handled by distributed, parallel, and federated databases?

• How is autonomy handled by distributed, parallel, and federated databases?

 

What is adaptive query processing

When adaptive query processing is beneficial in ddb

What are the situations in which adaptive query processing is beneficial