Unit 920 - Distributed Systems

1. Introduction to Distributed Systems

• Definition and scope of distributed systems
• Key characteristics: concurrency, lack of a global clock, independent failures
• Advantages: resource sharing, scalability, fault tolerance
• Challenges: synchronization, fault handling, consistency, security

2. Communication in Distributed Systems

• Communication models: message passing, remote procedure calls (RPC), remote method invocation (RMI)
• Communication protocols: TCP/IP, UDP, HTTP, gRPC
• Communication architectures: client-server, peer-to-peer
• Serialization and marshaling techniques

3. Distributed System Architectures

• Client-server architecture: roles, interaction patterns
• Peer-to-peer architecture: decentralized control, scalability
• Microservices architecture: modularity, service discovery, orchestration
• Cloud computing architecture: service models (IaaS, PaaS, SaaS), deployment models

4. Distributed Computing Paradigms

• Parallel computing: concepts, shared vs distributed memory
• Cluster computing: tightly coupled nodes, high availability
• Grid computing: resource sharing across administrative domains
• Cloud computing: on-demand resource provisioning, elasticity
• Applications, scalability, and fault tolerance considerations in each paradigm

5. Distributed File Systems

• Design goals: transparency, fault tolerance, scalability
• Hadoop Distributed File System (HDFS): architecture, data blocks, replication
• Network File System (NFS): protocol overview, stateless vs stateful design
• Ceph: object storage, CRUSH algorithm, scalability and fault tolerance
• Data replication, consistency models, access protocols

6. Distributed Data Management

• Distributed database systems: architecture, query processing, transaction management
• NoSQL databases: types (key-value, document, column-family, graph), use cases
• Data replication strategies: synchronous vs asynchronous replication
• Consistency models: strong, eventual, causal consistency
• CAP theorem and trade-offs

7. Fault Tolerance and Replication

• Fault types: crash, omission, timing, Byzantine
• Fault tolerance mechanisms: redundancy, checkpointing, rollback recovery
• Replication strategies: primary-backup, multi-primary
• Consistency models in replication
• Recovery techniques and consensus algorithms (e.g., Paxos, Raft)

8. Scalability and Load Balancing

• Scalability challenges: bottlenecks, resource contention
• Load balancing algorithms: round-robin, least connections, weighted distribution
• Horizontal scaling vs vertical scaling
• Auto-scaling and elasticity in cloud systems
• Performance metrics and monitoring

9. Security and Privacy in Distributed Systems

• Security threats: eavesdropping, spoofing, denial of service, insider threats
• Authentication mechanisms: passwords, certificates, multi-factor authentication
• Encryption techniques: symmetric, asymmetric, TLS/SSL
• Access control models: discretionary, mandatory, role-based access control
• Privacy concerns and data protection regulations

10. Case Studies in Distributed Systems

• Google's Bigtable: architecture, data model, scalability
• Amazon DynamoDB: design principles, eventual consistency, partitioning
• Apache Hadoop: ecosystem overview, MapReduce, HDFS
• Microsoft Azure: cloud services architecture, scalability, security features
• Performance characteristics and lessons learned