分布式系统期末复习重点.doc

chapter1:Introduction Definition of a Distributed Sys A distributed system is a collection of independent computers(Hardware) that appears to its users as a single coherent(一致的) system(Software) Four goals of DS Easily connect users to resources; Hide the fact that resources are distributed across a network; Openness(开放性); Scalability(可扩展性) Types of DS Distributed computing sys: Cluster computing sys, Grid computing sys, Transaction processing sys, Enterprise application integration chapter2:Architectures Centralized architectures The simplest organization that only have two types of machines: a client machine containing only the programs implementing the user-interface level; a server machine containing the rest, the programs implementing the processing and data level Decentralized architectures : Peer-to-Peer A P2P system is a distributed collection of peer nodes, each node is both a server and a client untructured P2P: Gnutella, Freenet, KaZaA structured P2P: Chord, CAN, Tapstry, Pastr, Kademlia, SkipNet, Cycloid P2P vs client-server Asymmetric(不对称) Symmetric Global knowledge Local knowledge Centralization Decentralization Single point of failure Robustness Limited scalability High scalability Expensive Low-cost Hybrid Architectures : Napster, BitTorrent central indexing; Each node registers to a central index; Each node acts as a client and central index as a server; Queries are performed to the central index; Retrieval is done from other ‘peer’nodes chapter3:Interprocess Communitions What is IPC (InterProcess Communications) the ability for separate, independent processes to communicate among themselves to collaborate on a task Event Synchronization two processes synchronize their operations: one side sends, then the other receives until all data has been sent and received Timeouts, Threading and Deadlocks blocking operations issued in the wrong sequence can cause deadlocks timeouts: may be used to set a maximum time period for blocking; it is provided b