Scribd
Upload
23 views3 pages

Distributed Systems QA Part1

The document outlines key characteristics and transparency requirements of distributed systems, including geographical distribution, message-based communication, and fault tolerance. It discusses properties of scalar time, termination detection, deadlock detection issues, mutual exclusion algorithm requirements, and rollback recovery message types. Additionally, it covers distributed file system requirements and compares whole file serving with whole file caching in AFS.

Uploaded by

FARAS A
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
23 views3 pages

Distributed Systems QA Part1

The document outlines key characteristics and transparency requirements of distributed systems, including geographical distribution, message-based communication, and fault tolerance. It discusses properties of scalar time, termination detection, deadlock detection issues, mutual exclusion algorithm requirements, and rollback recovery message types. Additionally, it covers distributed file system requirements and compares whole file serving with whole file caching in AFS.

Uploaded by

FARAS A
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Distributed Systems Q&A

1. Characteristics of Distributed Systems

- Geographical distribution

- No shared memory, communication via messages

- Concurrency among components

- No global clock

- Scalability, fault tolerance, and modularity

- Transparency in access, location, migration, replication, concurrency, and failure

2. Transparency Requirements of Distributed Systems

- Access Transparency

- Location Transparency

- Migration Transparency

- Relocation Transparency

- Replication Transparency

- Concurrency Transparency

- Failure Transparency

3. Basic Properties of Scalar Time

- Consistency: If ei -> ej, then C(ei) < C(ej)

- Total Ordering: Events totally ordered using scalar clocks with process ID

- Event Counting: Timestamp - 1 gives number of prior events

- No Strong Consistency: C(ei) < C(ej) does not imply ei -> ej

4. Termination Detection

Termination occurs when all processes are idle and no messages are in transit. Techniques include:

- Spanning-tree-based algorithms

- Snapshot-based detection

- Weight throwing algorithms


Distributed Systems Q&A

5. Issues in Deadlock Detection

- Detection: Uses Wait-for Graph (WFG)

- Correctness: Ensures progress and safety

- Resolution: Involves rollback or preemption

6. Requirements of Mutual Exclusion Algorithms

- Mutual Exclusion

- Progress

- Fairness

- Bounded Waiting

- Freedom from Deadlock and Starvation

7. Types of Messages in Rollback Recovery

- Checkpoint messages

- Rollback/Restart messages

- Log or Recovery info messages

8. No Orphans Consistency Condition

- Orphan processes are ones whose state can't be reconstructed

- No-orphan condition ensures recovery avoids such inconsistencies

9. Distributed File System Requirements

- Transparency (naming, location, replication, concurrency)

- Scalability

- Fault Tolerance

- Security, consistency, efficient data access

10. Whole File Serving vs. Whole File Caching in AFS


Distributed Systems Q&A

- Whole File Serving: Entire file is served each time

- Whole File Caching: File is cached after first access; future accesses are local

You might also like