Slide 1 – TitleSlide 1 – Distributed Database Collection of logically related databases. Data is stored at different network-connected sites. DDBMS manages the distributed data. Distribution is hidden from users. Slide 2 – Types & Architecture Types: Homogeneous: Same DBMS at all sites. Heterogeneous: Different DBMSs at different sites. Architecture: Client–Server Peer-to-Peer Multi-DBMS Slide 3 – Distributed Database Design Distributed design decides how and where data is stored. Main techniques: Replication Fragmentation Goals: Better reliability Faster access Reduced network load Slide 4 – Replication Replication stores copies of the same data at multiple sites. Advantages: Higher reliability Faster response Reduced network load Disadvantage: More storage and complex updates Slide 5 – Fragmentation Fragmentation divides a database into smaller fragments. Types: Horizontal: Divides rows. Vertical: Divides columns and retains the primary key. Hybrid: Combines horizontal and vertical fragmentation. Slide 6 – Data Transparency Data transparency hides distribution complexity from users. Types: Fragmentation Transparency Replication Transparency Location Transparency Users can access distributed data as if it were a single database. Slide 7 – Concurrency Control Concurrency means multiple transactions execute simultaneously. Concurrency control maintains correctness and consistency. Challenges: Multiple data copies Site failures Communication failures Network partition Distributed deadlock Distributed commit Slide 8 – Concurrency Techniques Locking Techniques: Distinguished Copy Primary Site Primary Site with Backup Primary Copy Timestamp Protocol: Transactions are processed according to timestamps. Slide 9 – Validation & Recovery Validation-Based Protocol: Read Phase Validation Phase Write Phase Distributed Recovery handles: Site failure Communication failure Network failure Transaction failure Slide 10 – Distributed Commit & Summary Distributed Commit: Records local effects permanently. Maintains local logs. Maintains consistency across sites. Overall Flow: Distribute Data → Design → Control Transactions → Recover → Maintain Consistency
Slide 1 – TitleSlide 1 – Distributed Database
Collection of logically related databases.
Data is stored at different network-connected sites.
DDBMS manages the distributed data.
Distribution is hidden from users.
Slide 2 – Types & Architecture
Types:
Homogeneous: Same DBMS at all sites.
Heterogeneous: Different DBMSs at different sites.
Architecture:
Client–Server
Peer-to-Peer
Multi-DBMS
Slide 3 – Distributed Database Design
Distributed design decides how and where data is stored.
Main techniques:
Replication
Fragmentation
Goals:
Better reliability
Faster access
Reduced network load
Slide 4 – Replication
Replication stores copies of the same data at multiple sites.
Advantages:
Higher reliability
Faster response
Reduced network load
Disadvantage:
More storage and complex updates
Slide 5 – Fragmentation
Fragmentation divides a database into smaller fragments.
Types:
Horizontal: Divides rows.
Vertical: Divides columns and retains the primary key.
Hybrid: Combines horizontal and vertical fragmentation.
Slide 6 – Data Transparency
Data transparency hides distribution complexity from users.
Types:
Fragmentation Transparency
Replication Transparency
Location Transparency
Users can access distributed data as if it were a single database.
Slide 7 – Concurrency Control
Concurrency means multiple transactions execute simultaneously.
Concurrency control maintains correctness and consistency.
Challenges:
Multiple data copies
Site failures
Communication failures
Network partition
Distributed deadlock
Distributed commit
Slide 8 – Concurrency Techniques
Locking Techniques:
Distinguished Copy
Primary Site
Primary Site with Backup
Primary Copy
Timestamp Protocol:
Transactions are processed according to timestamps.
Slide 9 – Validation & Recovery
Validation-Based Protocol:
Read Phase
Validation Phase
Write Phase
Distributed Recovery handles:
Site failure
Communication failure
Network failure
Transaction failure
Slide 10 – Distributed Commit & Summary
Distributed Commit:
Records local effects permanently.
Maintains local logs.
Maintains consistency across sites.
Overall Flow:
Distribute Data → Design → Control Transactions → Recover → Maintain Consistency
Created using ChatSlide
This overview covers essential concepts of Distributed Databases (DDBs) and Distributed Database Management Systems (DDBMS). It contrasts homogeneous and heterogeneous systems, delves into architectural designs, and emphasizes data transparency. The design section focuses on enhancing performance through replication and fragmentation strategies, ensuring a balance between storage and network load. Finally, it addresses control mechanisms, transaction validation, and recovery processes,...