Introduction-to-Distributed-Systems Introduction-to-Distributed-Systems.pdf
- 1. Distributed Systems Dr. Turkan Ahmed Khaleel Master's in Computer Engineering
- 2. What are Distributed Systems? Definition A Distributed System is a collection of independent computers that cooperate to achieve a common goal. These systems are designed to handle complex tasks that require distributed processing, storage, and communication. Importance Distributed systems are essential for modern applications and services. They enable scalability, high availability, and fault tolerance, making them suitable for large-scale data processing, online platforms, and complex business operations.
- 3. Key Characteristics of Distributed Systems Concurrency Multiple processes can execute simultaneously. In file-sharing systems, users can upload or download files concurrently. No Global Clock Lack of unified time across components makes coordination challenging. Financial trading systems require special mechanisms for coordinating simultaneous price changes in different regions. Independent Failures Components can fail independently without affecting the entire system. In distributed data storage, if one server fails, others continue to provide data.
- 4. Examples of Distributed Systems Web Search Engines Google uses distributed infrastructure, file systems, and storage to index billions of web pages. Their solutions include the Google File System (GFS) and Bigtable for fast access to massive datasets. Massively Multiplayer Online Games (MMOGs) These games support thousands of players simultaneously using client- server architecture and partitioning of the virtual world across different servers. Financial Trading Systems These systems process financial events in real-time using Complex Event Processing (CEP) and event- based distributed systems to analyze and distribute stock price changes quickly.
- 5. Trends in Distributed Systems Pervasive Networking Modern Internet consists of diverse networks, including wireless networks like Wi-Fi and Bluetooth, allowing devices to connect anywhere, anytime. Mobile and Ubiquitous Computing Mobile devices and embedded computers in everyday objects enable access to resources on the move and from anywhere. Cloud Computing Computing resources are provided as a service, offering flexibility, low cost, and scalability on demand.
- 6. Focus on Resource Sharing The main goal of distributed systems is sharing resources such as printers, files, and databases among users. This is often achieved through the client-server model. Client Requests access to resources. For example, a browser requests web pages from the server. Server Manages resources and provides them to clients. For instance, a web server provides web pages. Resource Shared assets like printers, files, or databases that clients access through the server.
- 7. Challenges in Distributed Systems 1 Security Protecting data from unauthorized access 2 Scalability Improving performance as users increase 3 Failure Handling Managing independent component failures 4 Concurrency Managing simultaneous resource access 5 Transparency Hiding system complexities from users Other challenges include heterogeneity, openness, and quality of service. Solutions involve middleware, APIs, encryption, caching, and synchronization mechanisms.
- 8. Real-World Examples of Distributed Systems Google Search Google's search engine utilizes a massive distributed system to index billions of web pages and deliver relevant results to users. Facebook Facebook's social network relies on a distributed system to handle billions of user interactions, store data, and manage content delivery. Blockchain Blockchain technology uses a distributed network of computers to secure and validate transactions, forming the basis for cryptocurrencies like Bitcoin.
- 9. Key Characteristics of Distributed Systems Transparency The system hides the complexity from from users. Users don't need to know know where resources are located or or how data is replicated. Scalability Can handle increased load by adding adding more nodes (computers) to the the system, increasing processing power power and capacity. Fault Tolerance The system continues working despite despite failures, ensuring high availability availability even if individual nodes experience problems. Concurrency Multiple processes can run simultaneously, allowing for parallel parallel execution and efficient resource utilization. Resource Sharing Nodes share resources such as files, databases, and computing power, enabling collaboration and efficient utilization of resources.
- 10. Challenges of Distributed Systems Concurrency Managing multiple requests and processes concurrently across different nodes can lead to race conditions, deadlocks, and inconsistencies. Data Consistency Ensuring that data is consistent across multiple nodes can be challenging, requiring mechanisms like distributed transactions and consensus protocols. Network Partitioning Network failures can isolate nodes, making communication difficult and requiring strategies like replication and fault tolerance.
- 11. Models of Distributed Systems Architectural Models • Client-Server: Clients request services from a central server (e.g., server (e.g., web browsers, web servers). • Peer-to-Peer (P2P): No central server, all nodes communicate communicate directly (e.g., BitTorrent, blockchain networks). networks). • Microservices Architecture: Applications are broken into smaller, smaller, independent services (e.g., Netflix streaming, payments). payments). Communication Models • Synchronous Communication: Sender waits for a response from response from the receiver. • Asynchronous Communication: Sender continues execution execution without waiting for a response. • Message Passing: Sending data between nodes (e.g., emails, emails, notifications). • Remote Procedure Call (RPC): Calling functions remotely, as if as if they were local (e.g., distributed database queries).
- 12. Common Distributed System Models Client-Server The most prevalent model. Clients request services from central servers. Examples include web applications, databases, and email services. Peer-to-Peer All nodes act as both client and server, communicating directly with each other. Examples include file-sharing networks and blockchain systems. Microservices An architectural approach where an application is decomposed into small, independent services. Each service runs independently and communicates through APIs. It promotes modularity, scalability, and flexibility.
- 13. Building a Basic Client- Server Model Step 1: Server Setup Create a Java server using sockets that listens for incoming connections and responds to client requests. Step 2: Client Setup Create a Java client that connects to the server, sends requests, and receives responses. Step 3: Communication Implement message exchange between the client and server using streams and data structures.
- 14. Case Study: The World Wide Web Web Components HTML for creating web pages, URLs for identifying resources, and HTTP for communication between browsers and servers. Dynamic Content Web pages generated dynamically using CGI programs or JavaScript, allowing for customized content based on user input. Web Services Allow programs to access web resources programmatically using XML and REST, enabling integration of diverse services and data sources.
- 15. The Future of Distributed Systems 1 Cloud Computing Distributed systems are becoming increasingly prevalent in cloud environments, enabling scalable and flexible solutions for various applications. 2 Edge Computing Edge computing brings processing and data storage closer to the users, enabling low-latency applications and reducing dependence on centralized data centers. 3 Serverless Architectures Serverless architectures allow developers to focus on code without managing servers, leveraging cloud-based platforms for scalable and efficient execution.
- 16. Summary of Distributed Systems Distributed systems are crucial for resource sharing and power various applications from web search to online gaming. They face key challenges including heterogeneity, security, scalability, and failure handling. The World Wide Web stands as a prime example of a successful distributed system, providing global access to information and services. It demonstrates how distributed systems can revolutionize information sharing and connectivity on a global scale.
- 17. Hands-on: Setting up a basic Client-Server model using model using Java Sockets Server Implementation The server listens for client connections and and responds with a message. It uses Java Java Sockets for network communication. communication. Client Implementation The client connects to the server, receives a receives a response message, and displays it displays it to the user. Running the Program Run the server first, followed by the client. client. The client will connect to the server server and receive a response message.