This preview shows pages 1–5. Sign up to view the full content.
1CSE120Principles of Operating SystemsProf Yuanyuan (YY) ZhouLecture 15: Distributed SystemsNov 18, 2010What is a Distributed System?Cooperating processes in a computer networkDegree of integration–Loose: Internet applications, email, web browsing–Medium: remote execution, remote file systems–Tight: distributed file systems11/17/2010CSE 1202BenefitsPerformance: parallelism across multiple nodes–Google file systems, BigTable, MapReduce, hadoop, etcReliability and fault tolerance–Redundancy–E.g.: Google search engineScalability by adding more nodes11/17/2010CSE 1203CSE 1204Clients and ServersThe prevalent model for structuring distributed computation is the client/server paradigmA serveris a program (or collection of programs) that provide a service(file server, name service, etc.)–The server may exist on one or more nodes–Often the node is called the server, too, which is confusingA clientis a program that uses the service–A client first bindsto the server (locates it and establishes a connection to it)–A client then sends requests, with data, to perform actions, and the servers sends responses, also with data11/17/2010
has intentionally blurred sections.
Sign up to view the full version.