Parallel Database Architecture Viva Questions

1. What are Parallel databases ?
   
   Parallel database systems use  multiple CPUs and Disks in Parallel to improve the performance through parallelization of various operations such as loading data ,building indexes, and evaluating queries. 

2. What are different types of Parallel databases architecture ?
   
   Shared Memory - Primary Memory is shared through interconnection Network
   
   Shared Disk - Hard Disk or Secondary Memory is shared  through interconnection Network
   
   Shared Nothing - Individual systems are connected through interconnection Network

3. In shared memory architecture which memory is shared primary or secondary ?
   
   Primary Memory

4.  What are advantages and disadvantages of shared memory ?
   
   Advantages -​

• It is closer to conventional machine ,
• Easy to program
• overhead is low.
• OS services are leveraged to utilize the additional CPUs.

Disadvantage:

• It leads to bottleneck problem
• Expensive to build
• It is less sensitive to partitioning

5. What are advantages and disadvantages of  shared disk  architecture ?

        Advantages:

• It is closer to conventional machine ,
• Easy to program
• overhead is low.
• OS services are leveraged to utilize the additional CPUs.

       Disadvantages:

• More interference.
• Increases N/W band width.
• Shared disk less sensitive to partitioning.

6. What are advantages and disadvantages of shared nothing  architecture ?
   Advantages:

• It provides linear scale up & linear speed up
• Shared nothing benefits from "good" partitioning
• Cheap to build

Disadvantage

• Hard to program.
• Addition of new nodes requires reorganizing.

7. Which type of architecture supports scaling ?
   
   Shared Nothing

8. Which architecture faces the problem of bottleneck ?
   
   Shared Memory

9. Which type of architecture has maximum fault tolerance ?
   
   Shared Disk

10. What is inter query parallelism ?
    
    When queries within a transaction or in multiple transactions  are processed in parallel to increase CPU performance and throughput time , it is called  interquery parallelism.eg We may have 10 people to execute 10 tasks so as to increase the performance and reduce the time taken to complete the job. 

11. What is intra query parallelism ?
    
    When a single query is broken down into subqueries and processed in parallel to increase CPU performance and throughput time , it is called  interaquery parallelism.eg We may break down 1 task in 10 different subtasks to get them executed faster.

12. What is Linear Speedup and sublinear speedup  in Parallel database ?
    
    Adding more CPU  increases the performance of the system  in Parallel databases up to a threshold value after that adding more CPU gives SubLinear Speedup. This scenario is similar to adding more people in a project. If we keep on adding people who are working on a single project then after a certain threshold value results may not be the same as expected as performance of the system may deteriorate due to overhead communications and other unforeseen issues.

13. What is Linear Scale up and sublinear scale up  in Parallel database ?
    
    Adding more nodes   increases the performance of the system  in Parallel databases up to a threshold value after that adding more nodes  gives SubLinear Speedup.

Various Search terms used are

Parallel Database Architecture Questions Answer

Parallel Database Viva  Questions Answer

Parallel Database Interview  Questions Answer

Parallel Database Short Questions Answer